Showing 1–50 of 72 results for author: Bradshaw, J

Nuclear Medicine Artificial Intelligence in Action: The Bethesda Report (AI Summit 2024)

Authors: Arman Rahmim, Tyler J. Bradshaw, Guido Davidzon, Joyita Dutta, Georges El Fakhri, Munir Ghesani, Nicolas A. Karakatsanis, Quanzheng Li, Chi Liu, Emilie Roncali, Babak Saboury, Tahir Yusufaly, Abhinav K. Jha

Abstract: The 2nd SNMMI Artificial Intelligence (AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD, on February 29 - March 1, 2024. Bringing together various community members and stakeholders, and following up on a prior successful 2022 AI Summit, the summit theme was: AI in Action. Six key topics included (i) an overview of prior and ongoing efforts by the AI task force, (ii) em… ▽ More The 2nd SNMMI Artificial Intelligence (AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD, on February 29 - March 1, 2024. Bringing together various community members and stakeholders, and following up on a prior successful 2022 AI Summit, the summit theme was: AI in Action. Six key topics included (i) an overview of prior and ongoing efforts by the AI task force, (ii) emerging needs and tools for computational nuclear oncology, (iii) new frontiers in large language and generative models, (iv) defining the value proposition for the use of AI in nuclear medicine, (v) open science including efforts for data and model repositories, and (vi) issues of reimbursement and funding. The primary efforts, findings, challenges, and next steps are summarized in this manuscript. △ Less

Submitted 3 June, 2024; originally announced June 2024.

Modeling Realistic Heating Profiles of Transition Region Hot Loops on the Sun: Evidence for Impulsive Heating and Non-equilibrium Ionization

Authors: Shah Mohammad Bahauddin, Stephen J Bradshaw

Abstract: The study examines the heating profile of hot solar transition region loops, particularly focusing on transient brightenings observed in IRIS 1400Å slit-jaw images. The findings challenge the adequacy of simplistic, singular heating mechanisms, revealing that the heating is temporally impulsive and requires a spatially complex profile with multiple heating scales. A forward modeling code is utiliz… ▽ More The study examines the heating profile of hot solar transition region loops, particularly focusing on transient brightenings observed in IRIS 1400Å slit-jaw images. The findings challenge the adequacy of simplistic, singular heating mechanisms, revealing that the heating is temporally impulsive and requires a spatially complex profile with multiple heating scales. A forward modeling code is utilized to generate synthetic IRIS emission spectra of these loops based on HYDRAD output, confirming that emitting ions are out of equilibrium. The modeling further indicates that density-dependent dielectronic recombination rates must be included to reproduce the observed line ratios. Collectively, this evidence substantiates that the loops are subject to impulsive heating and that the components of the transiently brightened plasma are driven far from thermal equilibrium. Heating events such as these are ubiquitous in the transition region and the analysis described above provides a robust observational diagnostic tool for characterizing the plasma. △ Less

Submitted 14 May, 2024; originally announced May 2024.

Comments: 11 Pages, 4 Figures, 3 Tables

Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-Aware Segmentation Network

Authors: Xin Tie, Muheon Shin, Changhee Lee, Scott B. Perlman, Zachary Huemann, Amy J. Weisman, Sharon M. Castellino, Kara M. Kelly, Kathleen M. McCarten, Adina L. Alazraki, Junjie Hu, Steve Y. Cho, Tyler J. Bradshaw

Abstract: $\textbf{Purpose}$: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. $\textbf{Materials and Metho… ▽ More $\textbf{Purpose}$: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. $\textbf{Materials and Methods}$: This retrospective study included baseline (PET1) and interim (PET2) PET/CT images from 297 patients enrolled in two Children's Oncology Group clinical trials (AHOD1331 and AHOD0831). LAS-Net incorporates longitudinal cross-attention, allowing relevant features from PET1 to inform the analysis of PET2. Model performance was evaluated using Dice coefficients for PET1 and detection F1 scores for PET2. Additionally, we extracted and compared quantitative PET metrics, including metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in PET1, as well as qPET and $Δ$SUVmax in PET2, against physician measurements. We quantified their agreement using Spearman's $ρ$ correlations and employed bootstrap resampling for statistical analysis. $\textbf{Results}$: LAS-Net detected residual lymphoma in PET2 with an F1 score of 0.606 (precision/recall: 0.615/0.600), outperforming all comparator methods (P<0.01). For baseline segmentation, LAS-Net achieved a mean Dice score of 0.772. In PET quantification, LAS-Net's measurements of qPET, $Δ$SUVmax, MTV and TLG were strongly correlated with physician measurements, with Spearman's $ρ$ of 0.78, 0.80, 0.93 and 0.96, respectively. The performance remained high, with a slight decrease, in an external testing cohort. $\textbf{Conclusion}$: LAS-Net achieved high performance in quantifying PET metrics across serial scans, highlighting the value of longitudinal awareness in evaluating multi-time-point imaging datasets. △ Less

Submitted 12 April, 2024; originally announced April 2024.

Comments: 6 figures, 4 tables in the main text

Beyond Major Product Prediction: Reproducing Reaction Mechanisms with Machine Learning Models Trained on a Large-Scale Mechanistic Dataset

Authors: Joonyoung F. Joung, Mun Hong Fong, Jihye Roh, Zhengkai Tu, John Bradshaw, Connor W. Coley

Abstract: Mechanistic understanding of organic reactions can facilitate reaction development, impurity prediction, and in principle, reaction discovery. While several machine learning models have sought to address the task of predicting reaction products, their extension to predicting reaction mechanisms has been impeded by the lack of a corresponding mechanistic dataset. In this study, we construct such a… ▽ More Mechanistic understanding of organic reactions can facilitate reaction development, impurity prediction, and in principle, reaction discovery. While several machine learning models have sought to address the task of predicting reaction products, their extension to predicting reaction mechanisms has been impeded by the lack of a corresponding mechanistic dataset. In this study, we construct such a dataset by imputing intermediates between experimentally reported reactants and products using expert reaction templates and train several machine learning models on the resulting dataset of 5,184,184 elementary steps. We explore the performance and capabilities of these models, focusing on their ability to predict reaction pathways and recapitulate the roles of catalysts and reagents. Additionally, we demonstrate the potential of mechanistic models in predicting impurities, often overlooked by conventional models. We conclude by evaluating the generalizability of mechanistic models to new reaction types, revealing challenges related to dataset diversity, consecutive predictions, and violations of atom conservation. △ Less

Submitted 7 March, 2024; originally announced March 2024.

Comments: 105 pages, 9 figures

Exploring Standing and Reflected Slow-mode Waves in Flaring Coronal Loops: A Parametric Study Using 2.5D MHD Modeling

Authors: Tongjiang Wang, Leon Ofman, Stephen J. Bradshaw

Abstract: Recent observations of reflected propagating and standing slow-mode waves in hot flaring coronal loops have spurred our investigation into their underlying excitation and dam** mechanisms. To understand these processes, we conduct 2.5D magnetohydrodynamic (MHD) simulations using an arcade active region model that includes a hot and dense loop. Our simulations allow for in-depth parametric invest… ▽ More Recent observations of reflected propagating and standing slow-mode waves in hot flaring coronal loops have spurred our investigation into their underlying excitation and dam** mechanisms. To understand these processes, we conduct 2.5D magnetohydrodynamic (MHD) simulations using an arcade active region model that includes a hot and dense loop. Our simulations allow for in-depth parametric investigations complementing and expanding our previous 3D MHD modeling results. We excite these waves in two distinct models as motivated by observations from the SDO/AIA. Model 1 incorporates classical compressive viscosity coefficient, while Model 2 adopts a 10-times enhanced viscosity coefficient. We find that: (1) Our 2.5D MHD simulations reinforce previous conclusions derived from 1D and 3D MHD models that significantly enhanced viscosity is crucial for the rapid excitation of standing slow waves with dam** times consistent with observations by Wang et al. (2015). (2) We uncover that nonlinearity in Model 1 delays the conversion of a reflected propagating wave into a standing wave. In contrast, Model 2 exhibits a much weak influence of nonlinearity. (3) Our results reveal that the transverse temperature structure holds more influence on wave behavior than the density structure. In Model 1, increased loop temperature contrast significantly enhances wave trap** within the structure, mitigating the impact of temperature-dependent viscous dam**. Conversely, in Model 2, the impact of temperature structure on wave behavior weakens in comparison to the effect of viscosity. (4) Model 1 displays evident nonlinear coupling to the fast and kink magnetoacoustic waves and pronounced wave leakage into the corona. However, analyzing three observed wave events by SDO/AIA aligns with Model 2 predictions, providing further support for the substantial viscosity increase. △ Less

Submitted 4 March, 2024; originally announced March 2024.

Comments: 44 pages, 19 figures, 3 tables, accepted for publication in Solar Physics

Journal ref: Solar Physics (2024) 299, 37

Anatomy and Physiology of Artificial Intelligence in PET Imaging

Authors: Tyler J. Bradshaw, Alan B. McMillan

Abstract: The influence of artificial intelligence (AI) within the field of nuclear medicine has been rapidly growing. Many researchers and clinicians are seeking to apply AI within PET, and clinicians will soon find themselves engaging with AI-based applications all along the chain of molecular imaging, from image reconstruction to enhanced reporting. This expanding presence of AI in PET imaging will resul… ▽ More The influence of artificial intelligence (AI) within the field of nuclear medicine has been rapidly growing. Many researchers and clinicians are seeking to apply AI within PET, and clinicians will soon find themselves engaging with AI-based applications all along the chain of molecular imaging, from image reconstruction to enhanced reporting. This expanding presence of AI in PET imaging will result in greater demand for educational resources for those unfamiliar with AI. The objective of this article to is provide an illustrated guide to the core principles of modern AI, with specific focus on aspects that are most likely to be encountered in PET imaging. We describe convolutional neural networks, algorithm training, and explain the components of the commonly used U-Net for segmentation and image synthesis. △ Less

Submitted 30 November, 2023; originally announced November 2023.

Journal ref: PET Clin; 16(4):471-482 (2021)

Automatic Personalized Impression Generation for PET Reports Using Large Language Models

Authors: Xin Tie, Muheon Shin, Ali Pirasteh, Nevein Ibrahim, Zachary Huemann, Sharon M. Castellino, Kara M. Kelly, John Garrett, Junjie Hu, Steve Y. Cho, Tyler J. Bradshaw

Abstract: In this study, we aimed to determine if fine-tuned large language models (LLMs) can generate accurate, personalized impressions for whole-body PET reports. Twelve language models were trained on a corpus of PET reports using the teacher-forcing algorithm, with the report findings as input and the clinical impressions as reference. An extra input token encodes the reading physician's identity, allo… ▽ More In this study, we aimed to determine if fine-tuned large language models (LLMs) can generate accurate, personalized impressions for whole-body PET reports. Twelve language models were trained on a corpus of PET reports using the teacher-forcing algorithm, with the report findings as input and the clinical impressions as reference. An extra input token encodes the reading physician's identity, allowing models to learn physician-specific reporting styles. Our corpus comprised 37,370 retrospective PET reports collected from our institution between 2010 and 2022. To identify the best LLM, 30 evaluation metrics were benchmarked against quality scores from two nuclear medicine (NM) physicians, with the most aligned metrics selecting the model for expert evaluation. In a subset of data, model-generated impressions and original clinical impressions were assessed by three NM physicians according to 6 quality dimensions (3-point scale) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. Bootstrap resampling was used for statistical analysis. Of all evaluation metrics, domain-adapted BARTScore and PEGASUSScore showed the highest Spearman's rank correlations (0.568 and 0.563) with physician preferences. Based on these metrics, the fine-tuned PEGASUS model was selected as the top LLM. When physicians reviewed PEGASUS-generated impressions in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08 out of 5. Physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P=0.41). In conclusion, personalized impressions generated by PEGASUS were clinically useful, highlighting its potential to expedite PET reporting. △ Less

Submitted 17 October, 2023; v1 submitted 18 September, 2023; originally announced September 2023.

Comments: 25 pages in total. 6 figures and 3 tables in the main body. The manuscript has been submitted to a journal for potential publication

Journal ref: J Digit Imaging. Inform. Med. (2024)

A Relationship Between Spin and Geometry

Authors: Peter T. J. Bradshaw

Abstract: In a recent paper, algebraic descriptions for all non-relativistic spins were derived by elementary means directly from the Lie algebra $\specialorthogonalliealgebra{3}$, and a connection between spin and the geometry of Euclidean three-space was drawn. However, the details of this relationship and the extent to which it can be developed by elementary means were not expounded. In this paper, we wi… ▽ More In a recent paper, algebraic descriptions for all non-relativistic spins were derived by elementary means directly from the Lie algebra $\specialorthogonalliealgebra{3}$, and a connection between spin and the geometry of Euclidean three-space was drawn. However, the details of this relationship and the extent to which it can be developed by elementary means were not expounded. In this paper, we will reveal the geometric content of the spin algebras by realising them within a novel, generalised form of Clifford-like algebra. In so doing, we will demonstrate a natural connection between spin and non-commutative geometry, and discuss the impact of this on the measurement of hypervolumes and on quantum mechanics. △ Less

Submitted 31 May, 2023; originally announced June 2023.

Comments: 12 pages, includes work presented at the 13th Annual Conference on Clifford Algebras and Their Applications in Mathematical Physics (ICCA13)

Prefix-Tree Decoding for Predicting Mass Spectra from Molecules

Authors: Samuel Goldman, John Bradshaw, Jiayi Xin, Connor W. Coley

Abstract: Computational predictions of mass spectra from molecules have enabled the discovery of clinically relevant metabolites. However, such predictive tools are still limited as they occupy one of two extremes, either operating (a) by fragmenting molecules combinatorially with overly rigid constraints on potential rearrangements and poor time complexity or (b) by decoding lossy and nonphysical discretiz… ▽ More Computational predictions of mass spectra from molecules have enabled the discovery of clinically relevant metabolites. However, such predictive tools are still limited as they occupy one of two extremes, either operating (a) by fragmenting molecules combinatorially with overly rigid constraints on potential rearrangements and poor time complexity or (b) by decoding lossy and nonphysical discretized spectra vectors. In this work, we use a new intermediate strategy for predicting mass spectra from molecules by treating mass spectra as sets of molecular formulae, which are themselves multisets of atoms. After first encoding an input molecular graph, we decode a set of molecular subformulae, each of which specify a predicted peak in the mass spectrum, the intensities of which are predicted by a second model. Our key insight is to overcome the combinatorial possibilities for molecular subformulae by decoding the formula set using a prefix tree structure, atom-type by atom-type, representing a general method for ordered multiset decoding. We show promising empirical results on mass spectra prediction tasks. △ Less

Submitted 3 December, 2023; v1 submitted 11 March, 2023; originally announced March 2023.

ConTEXTual Net: A Multimodal Vision-Language Model for Segmentation of Pneumothorax

Authors: Zachary Huemann, Xin Tie, Junjie Hu, Tyler J. Bradshaw

Abstract: Radiology narrative reports often describe characteristics of a patient's disease, including its location, size, and shape. Motivated by the recent success of multimodal learning, we hypothesized that this descriptive text could guide medical image analysis algorithms. We proposed a novel vision-language model, ConTEXTual Net, for the task of pneumothorax segmentation on chest radiographs. ConTEXT… ▽ More Radiology narrative reports often describe characteristics of a patient's disease, including its location, size, and shape. Motivated by the recent success of multimodal learning, we hypothesized that this descriptive text could guide medical image analysis algorithms. We proposed a novel vision-language model, ConTEXTual Net, for the task of pneumothorax segmentation on chest radiographs. ConTEXTual Net utilizes language features extracted from corresponding free-form radiology reports using a pre-trained language model. Cross-attention modules are designed to combine the intermediate output of each vision encoder layer and the text embeddings generated by the language model. ConTEXTual Net was trained on the CANDID-PTX dataset consisting of 3,196 positive cases of pneumothorax with segmentation annotations from 6 different physicians as well as clinical radiology reports. Using cross-validation, ConTEXTual Net achieved a Dice score of 0.716$\pm$0.016, which was similar to the degree of inter-reader variability (0.712$\pm$0.044) computed on a subset of the data. It outperformed both vision-only models (ResNet50 U-Net: 0.677$\pm$0.015 and GLoRIA: 0.686$\pm$0.014) and a competing vision-language model (LAVT: 0.706$\pm$0.009). Ablation studies confirmed that it was the text information that led to the performance gains. Additionally, we show that certain augmentation methods degraded ConTEXTual Net's segmentation performance by breaking the image-text concordance. We also evaluated the effects of using different language models and activation functions in the cross-attention module, highlighting the efficacy of our chosen architectural design. △ Less

Submitted 15 September, 2023; v1 submitted 2 March, 2023; originally announced March 2023.

Issues and Challenges in Applications of Artificial Intelligence to Nuclear Medicine -- The Bethesda Report (AI Summit 2022)

Authors: Arman Rahmim, Tyler J. Bradshaw, Irène Buvat, Joyita Dutta, Abhinav K. Jha, Paul E. Kinahan, Quanzheng Li, Chi Liu, Melissa D. McCradden, Babak Saboury, Eliot Siegel, John J. Sunderland, Richard L. Wahl

Abstract: The SNMMI Artificial Intelligence (SNMMI-AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD on March 21-22, 2022. It brought together various community members and stakeholders from academia, healthcare, industry, patient representatives, and government (NIH, FDA), and considered various key themes to envision and facilitate a bright future for routine, trustworthy use of… ▽ More The SNMMI Artificial Intelligence (SNMMI-AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD on March 21-22, 2022. It brought together various community members and stakeholders from academia, healthcare, industry, patient representatives, and government (NIH, FDA), and considered various key themes to envision and facilitate a bright future for routine, trustworthy use of AI in nuclear medicine. In what follows, essential issues, challenges, controversies and findings emphasized in the meeting are summarized. △ Less

Submitted 7 November, 2022; originally announced November 2022.

The ASAS-SN Bright Supernova Catalog -- V. 2018-2020

Authors: K. D. Neumann, T. W. -S. Holoien, C. S. Kochanek, K. Z. Stanek, P. J. Vallely, B. J. Shappee, J. L. Prieto, T. Pessi, T. Jayasinghe, J. Brimacombe, D. Bersier, E. Aydi, C. Basinger, J. F. Beacom, S. Bose, J. S. Brown, P. Chen, A. Clocchiatti, D. D. Desai, Subo Dong, E. Falco, S. Holmbo, N. Morrell, J. V. Shields, K. V. Sokolovsky , et al. (33 additional authors not shown)

Abstract: We catalog the 443 bright supernovae discovered by the All-Sky Automated Survey for Supernovae (ASAS-SN) in $2018-2020$ along with the 519 supernovae recovered by ASAS-SN and 516 additional $m_{peak}\leq18$ mag supernovae missed by ASAS-SN. Our statistical analysis focuses primarily on the 984 supernovae discovered or recovered in ASAS-SN $g$-band observations. The complete sample of 2427 ASAS-SN… ▽ More We catalog the 443 bright supernovae discovered by the All-Sky Automated Survey for Supernovae (ASAS-SN) in $2018-2020$ along with the 519 supernovae recovered by ASAS-SN and 516 additional $m_{peak}\leq18$ mag supernovae missed by ASAS-SN. Our statistical analysis focuses primarily on the 984 supernovae discovered or recovered in ASAS-SN $g$-band observations. The complete sample of 2427 ASAS-SN supernovae includes earlier $V$-band samples and unrecovered supernovae. For each supernova, we identify the host galaxy, its UV to mid-IR photometry, and the offset of the supernova from the center of the host. Updated light curves, redshifts, classifications, and host galaxy identifications supersede earlier results. With the increase of the limiting magnitude to $g\leq18$ mag, the ASAS-SN sample is roughly complete up to $m_{peak}=16.7$ mag and is $90\%$ complete for $m_{peak}\leq17.0$ mag. This is an increase from the $V$-band sample where it was roughly complete up to $m_{peak}=16.2$ mag and $70\%$ complete for $m_{peak}\leq17.0$ mag. △ Less

Submitted 24 February, 2023; v1 submitted 12 October, 2022; originally announced October 2022.

Comments: 14 pages, 7 figures, 4 tables. Updated to reflect changes made in the published version. Tables containing the catalog data presented in this submission are included in machine-readable format as ancillary files

Journal ref: MNRAS, 520, 4356 (2023)

Temperature and Differential Emission Measure Profiles in Turbulent Solar Active Region Loops

Authors: A. Gordon Emslie, Stephen J. Bradshaw

Abstract: We examine the temperature structure of static coronal active region loops in regimes where thermal conductive transport is driven by Coulomb collisions, by turbulent scattering, or by a combination of the two. (In the last case collisional scattering dominates the heat transport at lower levels in the loop where temperatures are low and densities are high, while turbulent scattering dominates the… ▽ More We examine the temperature structure of static coronal active region loops in regimes where thermal conductive transport is driven by Coulomb collisions, by turbulent scattering, or by a combination of the two. (In the last case collisional scattering dominates the heat transport at lower levels in the loop where temperatures are low and densities are high, while turbulent scattering dominates the heat transport at higher temperatures/lower densities.) Temperature profiles and their corresponding differential emission measure distributions are calculated and compared to observations, and earlier scaling laws relating the loop apex temperature and volumetric heating rate to the loop length and pressure are revisited. Results reveal very substantial changes, compared to the wholly collision-dominated case, to both the loop scaling laws and the temperature/density profiles along the loop. They also show that the well-known excess of differential emission measure at relatively low temperatures in the loop may be a consequence of for by the flatter temperature gradients (and so increased amount of material within a specified temperature range) that results from the predominance of turbulent scattering in the upper regions of the loop. △ Less

Submitted 3 October, 2022; originally announced October 2022.

An Algebraic Theory of Non-Relativistic Spin

Authors: Peter T. J. Bradshaw

Abstract: In this paper we present a new, elementary derivation of non-relativistic spin using exclusively real algebraic methods. To do this, we formulate a novel method to decompose the domain of a real endomorphism according to its algebraic properties. We reveal non-commutative multipole tensors as the primary physically meaningful observables of spin, and indicate that spin is fundamentally geometric i… ▽ More In this paper we present a new, elementary derivation of non-relativistic spin using exclusively real algebraic methods. To do this, we formulate a novel method to decompose the domain of a real endomorphism according to its algebraic properties. We reveal non-commutative multipole tensors as the primary physically meaningful observables of spin, and indicate that spin is fundamentally geometric in nature. In so doing, we demonstrate that neither dynamics nor complex numbers are essential to the fundamental description of spin. △ Less

Submitted 6 October, 2022; v1 submitted 5 July, 2022; originally announced July 2022.

Comments: 27 pages, 1 figure, 2 tables, v2 changes: Restructured introduction. Restructured arguments to use the step maps earlier. Added step-level and step-up in terms of multipoles. Fixed typos

The Dynamic Evolution of Solar Wind Streams Following Interchange Reconnection

Authors: Roger B. Scott, Stephen J. Bradshaw, Mark G. Linton

Abstract: Interchange reconnection is thought to play an important role in determining the dynamics and material composition of the slow solar wind that originates from near coronal hole boundaries. To explore the implications of this process we simulate the dynamic evolution of a solar wind stream along a newly-opened magnetic flux tube. The initial condition is composed of a piecewise continuous dynamic e… ▽ More Interchange reconnection is thought to play an important role in determining the dynamics and material composition of the slow solar wind that originates from near coronal hole boundaries. To explore the implications of this process we simulate the dynamic evolution of a solar wind stream along a newly-opened magnetic flux tube. The initial condition is composed of a piecewise continuous dynamic equilibrium in which the regions above and below the reconnection site are extracted from steady-state solutions along open and closed field lines. The initial discontinuity at the reconnection site is highly unstable and evolves as a Riemann problem, decomposing into an outward-propagating shock and inward-propagating rarefaction that eventually develop into a classic N-wave configuration. This configuration ultimately propagates into the heliosphere as a coherent structure and the entire system eventually settles to a quasi-steady wind solution. In addition to simulating the fluid evolution we also calculate the time-dependent non-equilibrium ionization of oxygen in real time in order to construct in situ diagnostics of the conditions near the reconnection site. This idealized description of the plasma dynamics along a newly-opened magnetic field line provides a baseline for predicting and interpreting the implications of interchange reconnection for the slow solar wind. Notably, the density and velocity within the expanding N-wave are generally enhanced over the ambient wind, as is the O7+/O6+ ionization ratio, which exhibits a discontinuity across the reconnection site that is transported by the flow and arrives later than the propagating N-wave. △ Less

Submitted 13 June, 2022; originally announced June 2022.

Validation of Monte Carlo Iodine-131 radiopharmaceutical dosimetry workflow using a 3D printed anthropomorphic head and neck phantom

Authors: David P. Adam, Joseph Grudzinski, Ian Bormett, Benjamin L. Cox, Ian R. Marsh, Tyler J. Bradshaw, Paul Harari, Bryan Bednarz

Abstract: Purpose: In this study, we present the creation of an anthropomorphic, head and neck, nuclear medicine phantom and its characterization for the validation of a Monte Carlo, SPECT image based, Iodine-131 RPT dosimetry workflow. Methods: 3D printing techniques were used to create the anthropomorphic phantom from a patient CT dataset. Three Iodine-131 SPECT/CT imaging studies were performed using a… ▽ More Purpose: In this study, we present the creation of an anthropomorphic, head and neck, nuclear medicine phantom and its characterization for the validation of a Monte Carlo, SPECT image based, Iodine-131 RPT dosimetry workflow. Methods: 3D printing techniques were used to create the anthropomorphic phantom from a patient CT dataset. Three Iodine-131 SPECT/CT imaging studies were performed using a homogeneous, Jaszczak, and an anthropomorphic phantom to quantify the SPECT images. The impact of collimator detector response (CDR) modeling and volume-based partial volume corrections (PVC) upon the absorbed dose was calculated using an image based, Geant4 Monte Carlo RPT dosimetry workflow and compared against a ground truth scenario. Finally, uncertainties were quantified in accordance with recent EANM guidelines. Results: The 3D printed anthropomorphic phantom was an accurate re-creation of patient anatomy including bone. The extrapolated Jaszczak recovery coefficients were greater than that of the 3D printed insert (~22.8 ml) for both the CDR and non-CDR cases. Utilizing Jaszczak phantom PVCs, the absorbed dose was underpredicted by 0.7% and 4.9% without and with CDR, respectively. Utilizing anthropomorphic phantom RCs overpredicted the absorbed dose by 3% both with and without CDR. All dosimetry scenarios that incorporated PVC were within the calculated uncertainty of the activity. The uncertainties in the cumulative activity ranged from 25.6% to 113% for Jaszczak spheres ranging in volume from 0.5 ml to 16 ml. Conclusion: The accuracy of Monte Carlo-based dosimetry for Iodine-131 RPT in head and neck cancer was validated with an anthropomorphic phantom. Future applications of the phantom could involve 3D printing and characterizing patient specific volumes for more personalized RPT dosimetry estimates. △ Less

Submitted 3 February, 2022; originally announced February 2022.

Comments: 30 pages, 11 figures, 4 tables

Local Latent Space Bayesian Optimization over Structured Inputs

Authors: Natalie Maus, Haydn T. Jones, Juston S. Moore, Matt J. Kusner, John Bradshaw, Jacob R. Gardner

Abstract: Bayesian optimization over the latent spaces of deep autoencoder models (DAEs) has recently emerged as a promising new approach for optimizing challenging black-box functions over structured, discrete, hard-to-enumerate search spaces (e.g., molecules). Here the DAE dramatically simplifies the search space by map** inputs into a continuous latent space where familiar Bayesian optimization tools c… ▽ More Bayesian optimization over the latent spaces of deep autoencoder models (DAEs) has recently emerged as a promising new approach for optimizing challenging black-box functions over structured, discrete, hard-to-enumerate search spaces (e.g., molecules). Here the DAE dramatically simplifies the search space by map** inputs into a continuous latent space where familiar Bayesian optimization tools can be more readily applied. Despite this simplification, the latent space typically remains high-dimensional. Thus, even with a well-suited latent space, these approaches do not necessarily provide a complete solution, but may rather shift the structured optimization problem to a high-dimensional one. In this paper, we propose LOL-BO, which adapts the notion of trust regions explored in recent work on high-dimensional Bayesian optimization to the structured setting. By reformulating the encoder to function as both an encoder for the DAE globally and as a deep kernel for the surrogate model within a trust region, we better align the notion of local optimization in the latent space with local optimization in the input space. LOL-BO achieves as much as 20 times improvement over state-of-the-art latent space Bayesian optimization methods across six real-world benchmarks, demonstrating that improvement in optimization strategies is as important as develo** better DAE models. △ Less

Submitted 22 February, 2023; v1 submitted 27 January, 2022; originally announced January 2022.

Static and dynamic solar coronal loops with cross-sectional area variations

Authors: P. J. Cargill, S. J. Bradshaw, J. A. Klimchuk, W. T. Barnes

Abstract: The Enthalpy Based Thermal Evolution of Loops (EBTEL) approximate model for static and dynamic coronal loops is developed to include the effect of a loop cross-sectional area which increases from the base of the transition region (TR) to the corona. The TR is defined as the part of a loop between the top of the chromosphere and the location where thermal conduction changes from an energy loss to a… ▽ More The Enthalpy Based Thermal Evolution of Loops (EBTEL) approximate model for static and dynamic coronal loops is developed to include the effect of a loop cross-sectional area which increases from the base of the transition region (TR) to the corona. The TR is defined as the part of a loop between the top of the chromosphere and the location where thermal conduction changes from an energy loss to an energy gain. There are significant differences from constant area loops due to the manner in which the reduced volume of the TR responds to conductive and enthalpy fluxes from the corona. For static loops with modest area variation the standard picture of loop energy balance is retained, with the corona and TR being primarily a balance between heating and conductive losses in the corona, and downward conduction and radiation to space in the TR. As the area at the loop apex increases, the TR becomes thicker and the density in TR and corona larger. For large apex areas, the coronal energy balance changes to one primarily between heating and radiation, with conduction playing an increasingly unimportant role, and the TR thickness becoming a significant fraction of the loop length. Approximate scaling laws are derived that give agreement with full numerical solutions for the density, but not the temperature. For non-uniform areas, dynamic loops have a higher peak temperature and are denser in the radiative cooling phase by of order 50% than the constant area case for the examples considered. They also show a final rapid cooling and draining once the temperature approaches 1 MK. Although the magnitude of the emission measure will be enhanced in the radiative phase, there is little change in the important observational diagnostic of its temperature dependence. △ Less

Submitted 17 November, 2021; originally announced November 2021.

Comments: 10 pages, 9 figures, published in MNRAS

Journal ref: Monthly Notices of the Royal Astronomical Society, 2021;, stab3163

Growth in Sumsets of Higher Convex Functions

Authors: Peter J. Bradshaw

Abstract: The main results of this paper concern growth in sums of a $k$-convex function $f$. Firstly, we streamline the proof of a growth result for $f(A)$ where $A$ has small additive doubling, and improve the bound by removing logarithmic factors. The result yields an optimal bound for \[ |2^k f(A) - (2^k-1)f(A)|. \] We also generalise a recent result of Hanson, Roche-Newton and Senger, by proving that… ▽ More The main results of this paper concern growth in sums of a $k$-convex function $f$. Firstly, we streamline the proof of a growth result for $f(A)$ where $A$ has small additive doubling, and improve the bound by removing logarithmic factors. The result yields an optimal bound for \[ |2^k f(A) - (2^k-1)f(A)|. \] We also generalise a recent result of Hanson, Roche-Newton and Senger, by proving that for any finite $A\subset \mathbb{R}$ \[ | 2^k f(sA-sA) - (2^k-1) f(sA-sA)| \gg_s |A|^{2s} \] where $s = \frac{k+1}{2}$. This allows us to prove that, given any natural number $s \in \mathbb{N}$, there exists $m = m(s)$ such that if $A \subset \mathbb{R}$, then \begin{equation}\label{conj A-Aus} |(sA-sA)^{(m)}| \gg_s |A|^{s}. \end{equation} This is progress towards a conjecture which states that the above inequality can be replaced with \[|(A-A)^{(m)}| \gg_s |A|^{s}.\] Develo** methods of Solymosi, and Bloom and Jones, we present some new sum-product type results in the complex numbers $\mathbb{C}$ and in the function field $\mathbb{F}_q((t^{-1}))$. △ Less

Submitted 5 November, 2021; originally announced November 2021.

Comments: 18 pages, comments welcome

Analysis of the first Genetic Engineering Attribution Challenge

Authors: Oliver M. Crook, Kelsey Lane Warmbrod, Greg Lipstein, Christine Chung, Christopher W. Bakerlee, T. Greg McKelvey Jr., Shelly R. Holland, Jacob L. Swett, Kevin M. Esvelt, Ethan C. Alley, William J. Bradshaw

Abstract: The ability to identify the designer of engineered biological sequences -- termed genetic engineering attribution (GEA) -- would help ensure due credit for biotechnological innovation, while holding designers accountable to the communities they affect. Here, we present the results of the first Genetic Engineering Attribution Challenge, a public data-science competition to advance GEA. Top-scoring… ▽ More The ability to identify the designer of engineered biological sequences -- termed genetic engineering attribution (GEA) -- would help ensure due credit for biotechnological innovation, while holding designers accountable to the communities they affect. Here, we present the results of the first Genetic Engineering Attribution Challenge, a public data-science competition to advance GEA. Top-scoring teams dramatically outperformed previous models at identifying the true lab-of-origin of engineered sequences, including an increase in top-1 and top-10 accuracy of 10 percentage points. A simple ensemble of prizewinning models further increased performance. New metrics, designed to assess a model's ability to confidently exclude candidate labs, also showed major improvements, especially for the ensemble. Most winning teams adopted CNN-based machine-learning approaches; however, one team achieved very high accuracy with an extremely fast neural-network-free approach. Future work, including future competitions, should further explore a wide diversity of approaches for bringing GEA technology into practical use. △ Less

Submitted 14 October, 2021; originally announced October 2021.

Comments: Main text: 11 pages, 4 figures, 37 references. Supplementary materials: 29 pages, 2 supplementary tables, 21 supplementary figures

AI-based data corrections for attenuation and scatter in PET and SPECT

Authors: Alan B. McMillan, Tyler J. Bradshaw

Abstract: Recent developments in artificial intelligence technology have enabled new developments that can improve attenuation and scatter correction in PET and SPECT. These technologies will enable the use of accurate and quantitative imaging without the need to acquire a CT image, greatly expanding the capability of PET/MRI, PET-only, and SPECT-only scanners. The use of AI to aid in scatter correction wil… ▽ More Recent developments in artificial intelligence technology have enabled new developments that can improve attenuation and scatter correction in PET and SPECT. These technologies will enable the use of accurate and quantitative imaging without the need to acquire a CT image, greatly expanding the capability of PET/MRI, PET-only, and SPECT-only scanners. The use of AI to aid in scatter correction will lead to improvements in image reconstruction speed, and improve patient throughput. This paper outlines the use of these new tools, surveys contemporary implementation, and discusses their limitations. △ Less

Submitted 27 August, 2021; originally announced August 2021.

Understanding Heating in Active Region Cores through Machine Learning II. Classifying Observations

Authors: W. T. Barnes, S. J. Bradshaw, N. M. Viall

Abstract: Constraining the frequency of energy deposition in magnetically-closed active region cores requires sophisticated hydrodynamic simulations of the coronal plasma and detailed forward modeling of the optically-thin line-of-sight integrated emission. However, understanding which set of model inputs best matches a set of observations is complicated by the need for any proposed heating model to simulta… ▽ More Constraining the frequency of energy deposition in magnetically-closed active region cores requires sophisticated hydrodynamic simulations of the coronal plasma and detailed forward modeling of the optically-thin line-of-sight integrated emission. However, understanding which set of model inputs best matches a set of observations is complicated by the need for any proposed heating model to simultaneously satisfy multiple observable constraints. In this paper, we train a random forest classification model on a set of forward-modeled observable quantities, namely the emission measure slope, the peak temperature of the emission measure distribution, and the time lag and maximum cross-correlation between multiple pairs of AIA channels. We then use our trained model to classify the heating frequency in every pixel of active region NOAA 1158 using the observed emission measure slopes, peak temperatures, time lags, and maximum cross-correlations and are able to map the heating frequency across the entire active region. We find that high-frequency heating dominates in the inner core of the active region while intermediate frequency dominates closer to the periphery of the active region. Additionally, we assess the importance of each observed quantity in our trained classification model and find that the emission measure slope is the dominant feature in deciding with which heating frequency a given pixel is most consistent. The technique presented here offers a very promising and widely applicable method for assessing observations in terms of detailed forward models given an arbitrary number of observable constraints. △ Less

Submitted 15 July, 2021; originally announced July 2021.

Comments: 29 pages, 8 figures, accepted for publication in The Astrophysical Journal

Higher Convexity and Iterated Second Moment Estimates

Authors: Peter J. Bradshaw, Brandon Hanson, Misha Rudnev

Abstract: We prove bounds for the number of solutions to $$a_1 + \dots + a_k = a_1' + \dots + a_k'$$ over $N$-element sets of reals, which are sufficiently convex or near-convex. A near-convex set will be the image of a set with small additive doubling under a convex function with sufficiently many strictly monotone derivatives. We show, roughly, that every time the number of terms in the equation is doub… ▽ More We prove bounds for the number of solutions to $$a_1 + \dots + a_k = a_1' + \dots + a_k'$$ over $N$-element sets of reals, which are sufficiently convex or near-convex. A near-convex set will be the image of a set with small additive doubling under a convex function with sufficiently many strictly monotone derivatives. We show, roughly, that every time the number of terms in the equation is doubled, an additional saving of $1$ in the exponent of the trivial bound $N^{2k-1}$ is made, starting from the trivial case $k=1$. In the context of near-convex sets we also provide explicit dependencies on the additive doubling parameters. Higher convexity is necessary for such bounds to hold, as evinced by sets of perfect powers of consecutive integers. We exploit these stronger assumptions using an idea of Garaev, rather than the ubiquitous Szemerédi-Trotter theorem, which has not been adapted in earlier results to embrace higher convexity. As an application we prove small improvements for the best known bounds for sumsets of convex sets under additional convexity assumptions. △ Less

Submitted 22 April, 2021; originally announced April 2021.

Comments: 22 pages

Expansion of Ultracold Neutral Plasmas with Exponentially Decaying Density Distributions

Authors: MacKenzie K. Warrens, Grant M. Gorman, Stephen J. Bradshaw, Thomas C. Killian

Abstract: We present a study of the expansion of an ultracold neutral plasma (UCNP) with an initial density distribution that decays exponentially in space, created by photoionizing atoms shortly after their release from a quadrupole (or biconic cusp) magnetic trap. A characteristic ion acoustic timescale is evident in the evolution of the plasma size and velocity, indicating the dynamics are reasonably wel… ▽ More We present a study of the expansion of an ultracold neutral plasma (UCNP) with an initial density distribution that decays exponentially in space, created by photoionizing atoms shortly after their release from a quadrupole (or biconic cusp) magnetic trap. A characteristic ion acoustic timescale is evident in the evolution of the plasma size and velocity, indicating the dynamics are reasonably well described by a model of hydrodynamic expansion of a quasi-neutral plasma. However, for low plasma density and high initial electron temperature, excess ion kinetic energy in the vicinity of the central density peak suggests significant local non-neutrality at early times. Observations are compared to the well-understood self-similar expansion of an UCNP with an initial Gaussian density distribution, and a similar scaling law describes the evolution of plasma size for both cases. △ Less

Submitted 28 December, 2020; originally announced December 2020.

Comments: 7 pages, 7 figures

Barking up the right tree: an approach to search over molecule synthesis DAGs

Authors: John Bradshaw, Brooks Paige, Matt J. Kusner, Marwin H. S. Segler, José Miguel Hernández-Lobato

Abstract: When designing new molecules with particular properties, it is not only important what to make but crucially how to make it. These instructions form a synthesis directed acyclic graph (DAG), describing how a large vocabulary of simple building blocks can be recursively combined through chemical reactions to create more complicated molecules of interest. In contrast, many current deep generative mo… ▽ More When designing new molecules with particular properties, it is not only important what to make but crucially how to make it. These instructions form a synthesis directed acyclic graph (DAG), describing how a large vocabulary of simple building blocks can be recursively combined through chemical reactions to create more complicated molecules of interest. In contrast, many current deep generative models for molecules ignore synthesizability. We therefore propose a deep generative model that better represents the real world process, by directly outputting molecule synthesis DAGs. We argue that this provides sensible inductive biases, ensuring that our model searches over the same chemical space that chemists would also have access to, as well as interpretability. We show that our approach is able to model chemical space well, producing a wide range of diverse molecules, and allows for unconstrained optimization of an inherently constrained problem: maximize certain chemical properties such that discovered molecules are synthesizable. △ Less

Submitted 21 December, 2020; originally announced December 2020.

Comments: To appear in Advances in Neural Information Processing Systems 2020

Magnetic Confinement of an Ultracold Neutral Plasma

Authors: G. M. Gorman, M. K. Warrens, S. J. Bradshaw, T. C. Killian

Abstract: We demonstrate magnetic confinement of an ultracold neutral plasma (UCNP) created at the null of a biconic cusp, or quadrupole magnetic field. Initially, the UCNP expands due to electron thermal pressure. As the plasma encounters stronger fields, expansion slows and the density distribution molds to the field. UCNP electrons are strongly magnetized over most of the plasma, while ion magnetization… ▽ More We demonstrate magnetic confinement of an ultracold neutral plasma (UCNP) created at the null of a biconic cusp, or quadrupole magnetic field. Initially, the UCNP expands due to electron thermal pressure. As the plasma encounters stronger fields, expansion slows and the density distribution molds to the field. UCNP electrons are strongly magnetized over most of the plasma, while ion magnetization is only significant at the boundaries. Observations suggest that electrons and ions are predominantly trapped by magnetic mirroring and ambipolar electric fields respectively. Confinement times approach 0.5 ms, while unmagnetized plasmas dissipate on a timescale of a few tens of microseconds. △ Less

Submitted 30 October, 2020; originally announced October 2020.

Comments: 5 pages, 4 figures

Journal ref: Phys. Rev. Lett. 126, 085002 (2021)

Stellar occultations enable milliarcsecond astrometry for Trans-Neptunian objects and Centaurs

Authors: F. L. Rommel, F. Braga-Ribas, J. Desmars, J. I. B. Camargo, J. L. Ortiz, B. Sicardy, R. Vieira-Martins, M. Assafin, P. Santos-Sanz, R. Duffard, E. Fernández-Valenzuela, J. Lecacheux, B. E. Morgado, G. Benedetti-Rossi, A. R. Gomes-Júnior, C. L. Pereira, D. Herald, W. Hanna, J. Bradshaw, N. Morales, J. Brimacombe, A. Burtovoi, T. Carruthers, J. R. de Barros, M. Fiori , et al. (44 additional authors not shown)

Abstract: Trans-Neptunian objects (TNOs) and Centaurs are remnants of our planetary system formation, and their physical properties have invaluable information for evolutionary theories. Stellar occultation is a ground-based method for studying these small bodies and has presented exciting results. These observations can provide precise profiles of the involved body, allowing an accurate determination of it… ▽ More Trans-Neptunian objects (TNOs) and Centaurs are remnants of our planetary system formation, and their physical properties have invaluable information for evolutionary theories. Stellar occultation is a ground-based method for studying these small bodies and has presented exciting results. These observations can provide precise profiles of the involved body, allowing an accurate determination of its size and shape. The goal is to show that even single-chord detections of TNOs allow us to measure their milliarcsecond astrometric positions in the reference frame of the Gaia second data release (DR2). Accurated ephemerides can then be generated, allowing predictions of stellar occultations with much higher reliability. We analyzed data from stellar occultations to obtain astrometric positions of the involved bodies. The events published before the Gaia era were updated so that the Gaia DR2 catalog is the reference. Previously determined sizes were used to calculate the position of the object center and its corresponding error with respect to the detected chord and the International Celestial Reference System (ICRS) propagated Gaia DR2 star position. We derive 37 precise astrometric positions for 19 TNOs and 4 Centaurs. Twenty-one of these events are presented here for the first time. Although about 68\% of our results are based on single-chord detection, most have intrinsic precision at the submilliarcsecond level. Lower limits on the diameter and shape constraints for a few bodies are also presented as valuable byproducts. Using the Gaia DR2 catalog, we show that even a single detection of a stellar occultation allows improving the object ephemeris significantly, which in turn enables predicting a future stellar occultation with high accuracy. Observational campaigns can be efficiently organized with this help, and may provide a full physical characterization of the involved object. △ Less

Submitted 23 October, 2020; originally announced October 2020.

Comments: 16 pages, 28 figures. The manuscript was accepted and is to be published

Journal ref: A&A 644, A40 (2020)

Scaling Laws for Dynamic Solar Loops

Authors: Stephen J. Bradshaw, A. Gordon Emslie

Abstract: The scaling laws which relate the peak temperature $T_M$ and volumetric heating rate $E_H$ to the pressure $P$ and length $L$ for static coronal loops were established over 40 years ago; they have proved to be of immense value in a wide range of studies. Here we extend these scaling laws to {\it dynamic} loops, where enthalpy flux becomes important to the energy balance, and study impulsive heatin… ▽ More The scaling laws which relate the peak temperature $T_M$ and volumetric heating rate $E_H$ to the pressure $P$ and length $L$ for static coronal loops were established over 40 years ago; they have proved to be of immense value in a wide range of studies. Here we extend these scaling laws to {\it dynamic} loops, where enthalpy flux becomes important to the energy balance, and study impulsive heating/filling characterized by upward enthalpy flows. We show that for collision-dominated thermal conduction, the functional dependencies of the scaling laws are the same as for the static case, when the radiative losses scale as $T^{-1/2}$, but with a different constant of proportionality that depends on the Mach number $M$ of the flow. The dependence on the Mach number is such that the scaling laws for low to moderate Mach number flows are almost indistinguishable from the static case. When thermal conduction is limited by turbulent processes, however, the much weaker dependence of the scattering mean free path (and hence thermal conduction coefficient) on temperature leads to a limiting Mach number for return enthalpy fluxes driven by thermal conduction between the corona and chromosphere. △ Less

Submitted 6 October, 2020; originally announced October 2020.

Comments: 14 pages, 3 figures

Critical Science Plan for the Daniel K. Inouye Solar Telescope (DKIST)

Authors: Mark P. Rast, Nazaret Bello González, Luis Bellot Rubio, Wenda Cao, Gianna Cauzzi, Edward DeLuca, Bart De Pontieu, Lyndsay Fletcher, Sarah E. Gibson, Philip G. Judge, Yukio Katsukawa, Maria D. Kazachenko, Elena Khomenko, Enrico Landi, Valentin Martínez Pillet, Gordon J. D. Petrie, Jiong Qiu, Laurel A. Rachmeler, Matthias Rempel, Wolfgang Schmidt, Eamon Scullion, Xudong Sun, Brian T. Welsch, Vincenzo Andretta, Patrick Antolin , et al. (62 additional authors not shown)

Abstract: The Daniel K. Inouye Solar Telescope (DKIST) will revolutionize our ability to measure, understand and model the basic physical processes that control the structure and dynamics of the Sun and its atmosphere. The first-light DKIST images, released publicly on 29 January 2020, only hint at the extraordinary capabilities which will accompany full commissioning of the five facility instruments. With… ▽ More The Daniel K. Inouye Solar Telescope (DKIST) will revolutionize our ability to measure, understand and model the basic physical processes that control the structure and dynamics of the Sun and its atmosphere. The first-light DKIST images, released publicly on 29 January 2020, only hint at the extraordinary capabilities which will accompany full commissioning of the five facility instruments. With this Critical Science Plan (CSP) we attempt to anticipate some of what those capabilities will enable, providing a snapshot of some of the scientific pursuits that the Daniel K. Inouye Solar Telescope hopes to engage as start-of-operations nears. The work builds on the combined contributions of the DKIST Science Working Group (SWG) and CSP Community members, who generously shared their experiences, plans, knowledge and dreams. Discussion is primarily focused on those issues to which DKIST will uniquely contribute. △ Less

Submitted 20 August, 2020; v1 submitted 18 August, 2020; originally announced August 2020.

Electron Beams Cannot Directly Produce Coronal Rain

Authors: Jeffrey W. Reep, Patrick Antolin, Stephen J. Bradshaw

Abstract: Coronal rain is ubiquitous in flare loops, forming shortly after the onset of the solar flare. Rain is thought to be caused by a thermal instability, a localized runaway cooling of material in the corona. The models that demonstrate this require extremely long duration heating on the order of the radiative cooling time, localized near the footpoints of the loops. In flares, electron beams are thou… ▽ More Coronal rain is ubiquitous in flare loops, forming shortly after the onset of the solar flare. Rain is thought to be caused by a thermal instability, a localized runaway cooling of material in the corona. The models that demonstrate this require extremely long duration heating on the order of the radiative cooling time, localized near the footpoints of the loops. In flares, electron beams are thought to be the primary energy transport mechanism, driving strong footpoint heating during the impulsive phase that causes evaporation, filling and heating flare loops. Electron beams, however, do not act for a long period of time, and even supposing that they did, their heating would not remain localized at the footpoints. With a series of numerical experiments, we show directly that these two issues mean that electron beams are incapable of causing the formation of rain in flare loops. This result suggests that either there is another mechanism acting in flare loops responsible for rain, or that the modeling of the cooling of flare loops is somehow deficient. To adequately describe flares, the standard model must address this issue to account for the presence of coronal rain. △ Less

Submitted 18 February, 2020; originally announced February 2020.

Comments: Published in ApJ

Journal ref: ApJ, Vol. 890, pp.100, 2020

Modelling the solar transition region using an adaptive conduction method

Authors: C. D. Johnston, P. J. Cargill, A. W. Hood, I. De Moortel, S. J. Bradshaw, A. C. Vaseekar

Abstract: Modelling the solar Transition Region with the use of an Adaptive Conduction (TRAC) method permits fast and accurate numerical solutions of the field-aligned hydrodynamic equations, capturing the enthalpy exchange between the corona and transition region, when the corona undergoes impulsive heating. The TRAC method eliminates the need for highly resolved numerical grids in the transition region an… ▽ More Modelling the solar Transition Region with the use of an Adaptive Conduction (TRAC) method permits fast and accurate numerical solutions of the field-aligned hydrodynamic equations, capturing the enthalpy exchange between the corona and transition region, when the corona undergoes impulsive heating. The TRAC method eliminates the need for highly resolved numerical grids in the transition region and the commensurate very short time steps that are required for numerical stability. When employed with coarse spatial resolutions, typically achieved in multi-dimensional magnetohydrodynamic codes, the errors at peak density are less than 5% and the computation time is three orders of magnitude faster than fully resolved field-aligned models. This paper presents further examples that demonstrate the versatility and robustness of the method over a range of heating events, including impulsive and quasi-steady footpoint heating. A detailed analytical assessment of the TRAC method is also presented, showing that the approach works through all phases of an impulsive heating event because (i) the total radiative losses and (ii) the total heating when integrated over the transition region are both preserved at all temperatures under the broadening modifications of the method. The results from the numerical simulations complement this conclusion. △ Less

Submitted 28 February, 2020; v1 submitted 5 February, 2020; originally announced February 2020.

Comments: 20 pages, 9 figures, 1 table, accepted for publication in A&A

Journal ref: A&A 635, A168 (2020)

Snake graphs and their characteristic polynomials

Authors: James P. Bradshaw, Philipp Lampe, Dusan Ziga

Abstract: The aim of the article is to understand the combinatorics of snake graphs by means of linear algebra. In particular, we apply Kasteleyn's and Temperley--Fisher's ideas about spectral properties of weighted adjacency matrices of planar bipartite graphs to snake graphs. First we focus on snake graphs whose set of turning vertices is monochromatic. We provide recursive sequences to compute the charac… ▽ More The aim of the article is to understand the combinatorics of snake graphs by means of linear algebra. In particular, we apply Kasteleyn's and Temperley--Fisher's ideas about spectral properties of weighted adjacency matrices of planar bipartite graphs to snake graphs. First we focus on snake graphs whose set of turning vertices is monochromatic. We provide recursive sequences to compute the characteristic polynomials; they are indexed by the upper or the lower boundary of the graph and are determined by a neighbour count. As an application, we compute the characteristic polynomials for L-shaped snake graphs and staircases in terms of Fibonacci product polynomials. Next, we introduce a method to compute the characteristic polynomials as convergents of continued fractions. Finally, we show how to transform a snake graph with turning vertices of two colours into a graph with the same number of perfect matchings to which we can apply the results above. △ Less

Submitted 25 October, 2019; originally announced October 2019.

Comments: 32 pages, 7 figures

MSC Class: 05C50 (Primary); 11Y65 (Secondary)

Solar Active Region Heating Diagnostics from High Temperature Emission using the Marshall Grazing Incidence X-ray Spectrometer (MaGIXS)

Authors: P. S. Athiray, Amy R. Winebarger, Will T. Barnes, Stephen J. Bradshaw, Sabrina Savage, Harry P. Warren, Ken Kobayashi, Patrick Champey, Leon Golub, Lindsay Glesener

Abstract: The relative amount of high temperature plasma has been found to be a useful diagnostic to determine the frequency of coronal heating on sub-resolution structures. When the loops are infrequently heated, a broad emission measure (EM) over a wider range of temperatures is expected. A narrower EM is expected for high frequency heating where the loops are closer to equilibrium. The soft X-ray spectru… ▽ More The relative amount of high temperature plasma has been found to be a useful diagnostic to determine the frequency of coronal heating on sub-resolution structures. When the loops are infrequently heated, a broad emission measure (EM) over a wider range of temperatures is expected. A narrower EM is expected for high frequency heating where the loops are closer to equilibrium. The soft X-ray spectrum contains many spectral lines that provide high temperature diagnostics, including lines from Fe XVII-XIX. This region of the solar spectrum will be observed by the Marshall Grazing Incidence Spectrometer (MaGIXS) in 2020. In this paper, we derive the expected spectral lines intensity in MaGIXS to varying amounts of high temperature plasma to demonstrate that a simple line ratio of these provides a powerful diagnostic to determine the heating frequency. Similarly, we examine ratios of AIA channel intensities, filter ratios from a XRT, and energy bands from the FOXSI sounding rocket to determine their sensitivity to this parameter. We find that both FOXSI and MaGIXS provide good diagnostic capability for high-temperature plasma. We then compare the predicted line ratios to the output of a numerical model and confirm the MaGIXS ratios provide an excellent diagnostic for heating frequency. △ Less

Submitted 5 September, 2019; originally announced September 2019.

Comments: Accepted for publication in ApJ

Journal ref: The Astrophysical Journal, 2019, 884 24

High-velocity outflows in massive post-starburst galaxies at z > 1

Authors: David T. Maltby, Omar Almaini, Ross J. McLure, Vivienne Wild, James Dunlop, Kate Rowlands, William G. Hartley, Nina A. Hatch, Miguel Socolovsky, Aaron Wilkinson, Ricardo Amorin, Emma J. Bradshaw, Adam C. Carnall, Marco Castellano, Andrea Cimatti, Giovanni Cresci, Fergus Cullen, Stephane De Barros, Fabio Fontanot, Bianca Garilli, Anton M. Koekemoer, Derek J. McLeod, Laura Pentericci, Margherita Talia

Abstract: We investigate the prevalence of galactic-scale outflows in post-starburst (PSB) galaxies at high redshift ($1 < z < 1.4$), using the deep optical spectra available in the UKIDSS Ultra Deep Survey (UDS). We use a sample of $\sim40$ spectroscopically confirmed PSBs, recently identified in the UDS field, and perform a stacking analysis in order to analyse the structure of strong interstellar absorpt… ▽ More We investigate the prevalence of galactic-scale outflows in post-starburst (PSB) galaxies at high redshift ($1 < z < 1.4$), using the deep optical spectra available in the UKIDSS Ultra Deep Survey (UDS). We use a sample of $\sim40$ spectroscopically confirmed PSBs, recently identified in the UDS field, and perform a stacking analysis in order to analyse the structure of strong interstellar absorption features such as Mg ii ($\lambda2800$ Ang.). We find that for massive ($M_* > 10^{10}\rm\,M_{\odot}$) PSBs at $z > 1$, there is clear evidence for a strong blue-shifted component to the Mg ii absorption feature, indicative of high-velocity outflows ($v_{\rm out}\sim1150\pm160\rm\,km\,s^{-1}$) in the interstellar medium. We conclude that such outflows are typical in massive PSBs at this epoch, and potentially represent the residual signature of a feedback process that quenched these galaxies. Using full spectral fitting, we also obtain a typical stellar velocity dispersion $σ_*$ for these PSBs of $\sim200\rm\,km\,s^{-1}$, which confirms they are intrinsically massive in nature (dynamical mass $M_{\rm d}\sim10^{11}\rm\,M_{\odot}$). Given that these high-$z$ PSBs are also exceptionally compact ($r_{\rm e}\sim1$--$2\rm\,kpc$) and spheroidal (Sersic index $n\sim3$), we propose that the outflowing winds may have been launched during a recent compaction event (e.g. major merger or disc collapse) that triggered either a centralised starburst or active galactic nuclei (AGN) activity. Finally, we find no evidence for AGN signatures in the optical spectra of these PSBs, suggesting they were either quenched by stellar feedback from the starburst itself, or that if AGN feedback is responsible, the AGN episode that triggered quenching does not linger into the post-starburst phase. △ Less

Submitted 7 August, 2019; originally announced August 2019.

Comments: 13 pages, 6 figures, accepted for publication in MNRAS

A Model to Search for Synthesizable Molecules

Authors: John Bradshaw, Brooks Paige, Matt J. Kusner, Marwin H. S. Segler, José Miguel Hernández-Lobato

Abstract: Deep generative models are able to suggest new organic molecules by generating strings, trees, and graphs representing their structure. While such models allow one to generate molecules with desirable properties, they give no guarantees that the molecules can actually be synthesized in practice. We propose a new molecule generation model, mirroring a more realistic real-world process, where (a) re… ▽ More Deep generative models are able to suggest new organic molecules by generating strings, trees, and graphs representing their structure. While such models allow one to generate molecules with desirable properties, they give no guarantees that the molecules can actually be synthesized in practice. We propose a new molecule generation model, mirroring a more realistic real-world process, where (a) reactants are selected, and (b) combined to form more complex molecules. More specifically, our generative model proposes a bag of initial reactants (selected from a pool of commercially-available molecules) and uses a reaction model to predict how they react together to generate new molecules. We first show that the model can generate diverse, valid and unique molecules due to the useful inductive biases of modeling reactions. Furthermore, our model allows chemists to interrogate not only the properties of the generated molecules but also the feasibility of the synthesis routes. We conclude by using our model to solve retrosynthesis problems, predicting a set of reactants that can produce a target product. △ Less

Submitted 4 December, 2019; v1 submitted 12 June, 2019; originally announced June 2019.

Comments: To appear in Advances in Neural Information Processing Systems 2019

Understanding Heating in Active Region Cores through Machine Learning I. Numerical Modeling and Predicted Observables

Authors: W. T. Barnes, S. J. Bradshaw, N. M. Viall

Abstract: To adequately constrain the frequency of energy deposition in active region cores in the solar corona, systematic comparisons between detailed models and observational data are needed. In this paper, we describe a pipeline for forward modeling active region emission using magnetic field extrapolations and field-aligned hydrodynamic models. We use this pipeline to predict time-dependent emission fr… ▽ More To adequately constrain the frequency of energy deposition in active region cores in the solar corona, systematic comparisons between detailed models and observational data are needed. In this paper, we describe a pipeline for forward modeling active region emission using magnetic field extrapolations and field-aligned hydrodynamic models. We use this pipeline to predict time-dependent emission from active region NOAA 1158 as observed by SDO/AIA for low-, intermediate-, and high-frequency nanoflares. In each pixel of our predicted multi-wavelength, time-dependent images, we compute two commonly-used diagnostics: the emission measure slope and the time lag. We find that signatures of the heating frequency persist in both of these diagnostics. In particular, our results show that the distribution of emission measure slopes narrows and the mean decreases with decreasing heating frequency and that the range of emission measure slopes is consistent with past observational and modeling work. Furthermore, we find that the time lag becomes increasingly spatially coherent with decreasing heating frequency while the distribution of time lags across the whole active region becomes more broad with increasing heating frequency. In a follow up paper, we train a random forest classifier on these predicted diagnostics and use this model to classify real AIA observations of NOAA 1158 in terms of the underlying heating frequency. △ Less

Submitted 7 June, 2019; originally announced June 2019.

Comments: 32 pages, 12 figures, accepted for publication in The Astrophysical Journal

Coronal Loop Scaling Laws for Various Forms of Parallel Heat Conduction

Authors: Stephen J. Bradshaw, A. Gordon Emslie, Nicolas H. Bian, Eduard P. Kontar

Abstract: The solar atmosphere is dominated by loops of magnetic flux which connect the multi-million-degree corona to the much cooler chromosphere. The temperature and density structure of quasi-static loops is determined by the continuous flow of energy from the hot corona to the lower solar atmosphere. Loop scaling laws provide relationships between global properties of the loop (such as peak temperature… ▽ More The solar atmosphere is dominated by loops of magnetic flux which connect the multi-million-degree corona to the much cooler chromosphere. The temperature and density structure of quasi-static loops is determined by the continuous flow of energy from the hot corona to the lower solar atmosphere. Loop scaling laws provide relationships between global properties of the loop (such as peak temperature, pressure, and length); they follow from the physical variable dependencies of various terms in the energy equation, and hence the form of the loop scaling law provides insight into the key physics that controls the loop structure. Traditionally, scaling laws have been derived under the assumption of collision-dominated thermal conduction. Here we examine the impact of different regimes of thermal conduction -- collision-dominated, turbulence-dominated, and free-streaming -- on the form of the scaling laws relating the loop temperature and heating rate to its pressure and half-length. We show that the scaling laws for turbulence-dominated conduction are fundamentally different than those for collision-dominated and free-streaming conduction, inasmuch as the form of the scaling laws now depend primarily on conditions at the low-temperature, rather than high-temperature, part of the loop. We also establish regimes in temperature and density space in which each of the applicable scaling laws prevail. △ Less

Submitted 7 June, 2019; originally announced June 2019.

The effects of numerical resolution, heating timescales and background heating on thermal non-equilibrium in coronal loops

Authors: C. D. Johnston, P. J. Cargill, P. Antolin, A. W. Hood, I. De Moortel, S. J. Bradshaw

Abstract: Thermal non-equilibrium (TNE) is believed to be a potentially important process in understanding some properties of the magnetically closed solar corona. Through one-dimensional hydrodynamic models, this paper addresses the importance of the numerical spatial resolution, footpoint heating timescales and background heating on TNE. Inadequate transition region (TR) resolution can lead to significant… ▽ More Thermal non-equilibrium (TNE) is believed to be a potentially important process in understanding some properties of the magnetically closed solar corona. Through one-dimensional hydrodynamic models, this paper addresses the importance of the numerical spatial resolution, footpoint heating timescales and background heating on TNE. Inadequate transition region (TR) resolution can lead to significant discrepancies in TNE cycle behaviour, with TNE being suppressed in under-resolved loops. A convergence on the periodicity and plasma properties associated with TNE required spatial resolutions of less than 2 km for a loop of length 180 Mm. These numerical problems can be resolved using an approximate method that models the TR as a discontinuity using a jump condition, as proposed by Johnston et al. (2017a,b). The resolution requirements (and so computational cost) are greatly reduced while retaining good agreement with fully resolved results. Using this approximate method we (i) identify different regimes for the response of coronal loops to time-dependent footpoint heating including one where TNE does not arise and (ii) demonstrate that TNE in a loop with footpoint heating is suppressed unless the background heating is sufficiently small. The implications for the generality of TNE are discussed. △ Less

Submitted 15 April, 2019; originally announced April 2019.

Comments: 15 pages, 10 figures, 1 table, accepted for publication in A&A

Journal ref: A&A 625, A149 (2019)

A Fast and Accurate Method to Capture the Solar Corona/Transition Region Enthalpy Exchange

Authors: C. D. Johnston, S. J. Bradshaw

Abstract: The brightness of the emission from coronal loops in the solar atmosphere is strongly dependent on the temperature and density of the confined plasma. After a release of energy, these loops undergo a heating and upflow phase, followed by a cooling and downflow cycle. Throughout, there are significant variations in the properties of the coronal plasma. In particular, the increased coronal temperatu… ▽ More The brightness of the emission from coronal loops in the solar atmosphere is strongly dependent on the temperature and density of the confined plasma. After a release of energy, these loops undergo a heating and upflow phase, followed by a cooling and downflow cycle. Throughout, there are significant variations in the properties of the coronal plasma. In particular, the increased coronal temperature leads to an excess downward heat flux that the transition region is unable to radiate. This generates an enthalpy flux from the transition region to the corona, increasing the coronal density. The enthalpy exchange is highly sensitive to the transition region resolution in numerical simulations. With a numerically under-resolved transition region, major errors occur in simulating the coronal density evolution and, thus, the predicted loop emission. This Letter presents a new method that addresses the difficulty of obtaining the correct interaction between the corona and corona/chromosphere interface. In the transition region, an adaptive thermal conduction approach is used that broadens any unresolved parts of the atmosphere. We show that this approach, referred to as TRAC, successfully removes the influence of numerical resolution on the coronal density response to heating while maintaining high levels of agreement with fully resolved models. When employed with coarse spatial resolutions, typically achieved in multi-dimensional MHD codes, the peak density errors are less than 3% and the computation time is three orders of magnitude faster than fully resolved field-aligned models. The advantages of using TRAC in field-aligned hydrodynamic and multi-dimensional magnetohydrodynamic simulations are discussed. △ Less

Submitted 4 March, 2019; originally announced March 2019.

Comments: 8 pages, 3 figures, accepted for publication in ApJL

Quantifying the Influence of Key Physical Processes on the Formation of Emission Lines Observed by IRIS: I. Non-Equilibrium Ionization and Density-Dependent Rates

Authors: Stephen J. Bradshaw, Paola Testa

Abstract: In the work described here we investigate atomic processes leading to the formation of emission lines within the IRIS wavelength range at temperatures near $10^5$~K. We focus on (1) non-equilibrium and (2) density-dependent effects influencing the formation and radiative properties of S IV and O IV. These two effects have significant impacts on spectroscopic diagnostic measurements of quantities a… ▽ More In the work described here we investigate atomic processes leading to the formation of emission lines within the IRIS wavelength range at temperatures near $10^5$~K. We focus on (1) non-equilibrium and (2) density-dependent effects influencing the formation and radiative properties of S IV and O IV. These two effects have significant impacts on spectroscopic diagnostic measurements of quantities associated with the plasma that emission lines from S IV and O IV provide. We demonstrate this by examining nanoflare-based coronal heating to determine what the detectable signatures are in transition region emission. A detailed comparison between predictions from numerical experiments and several sets of observational data is presented to show how one can ascertain when non-equilibrium ionization and/or density-dependent atomic processes are important for diagnosing nanoflare properties, the magnitude of their contribution, and what information can be reliably extracted from the spectral data. Our key findings are the following. (1) The S/O intensity ratio is a powerful diagnostic of non-equilibrium ionization. (2) Non-equilibrium ionization has a strong effect on the observed line intensities even in the case of relatively weak nanoflare heating. (3) The density-dependence of atomic rate coefficients is only important when the ion population is out of equilibrium. (4) In the sample of active regions we examined, weak nanoflares coupled with non-equilibrium ionization and density-dependent atomic rates were required to explain the observed properties (e.g. the S/O intensity ratios). (5) Enhanced S/O intensity ratios cannot be due solely to the heating strength and must depend on other processes (e.g. heating frequency, non-Maxwellian distributions). △ Less

Submitted 12 January, 2019; originally announced January 2019.

Comments: 26 pages, 13 figures

Hard X-Ray Constraints on Small-Scale Coronal Heating Events

Authors: Andrew J. Marsh, David M. Smith, Lindsay Glesener, James A. Klimchuk, Stephen J. Bradshaw, Juliana Vievering, Iain G. Hannah, Steven Christe, Shin-nosuke Ishikawa, Sam Krucker

Abstract: Much evidence suggests that the solar corona is heated impulsively, meaning that nanoflares may be ubiquitous in quiet and active regions (ARs). Hard X-ray (HXR) observations with unprecedented sensitivity $>$3~keV are now enabled by focusing instruments. We analyzed data from the \textit{Focusing Optics X-ray Solar Imager (FOXSI)} rocket and the \textit{Nuclear Spectroscopic Telescope Array (NuST… ▽ More Much evidence suggests that the solar corona is heated impulsively, meaning that nanoflares may be ubiquitous in quiet and active regions (ARs). Hard X-ray (HXR) observations with unprecedented sensitivity $>$3~keV are now enabled by focusing instruments. We analyzed data from the \textit{Focusing Optics X-ray Solar Imager (FOXSI)} rocket and the \textit{Nuclear Spectroscopic Telescope Array (NuSTAR)} spacecraft to constrain properties of AR nanoflares simulated by the EBTEL field-line-averaged hydrodynamics code. We generated model X-ray spectra by computing differential emission measures for homogeneous nanoflare sequences with heating amplitudes $H_0$, durations $τ$, delay times between events $t_N$, and filling factors $f$. The single quiescent AR observed by \textit{FOXSI-2} on 2014 December 11 is well fit by nanoflare sequences with heating amplitudes 0.02 erg cm$^{-3}$ s$^{-1}$ $<$ $H_0$ $<$ 13 erg cm$^{-3}$ s$^{-1}$ and a wide range of delay times and durations. We exclude delays between events shorter than $\sim$900 s at the 90\% confidence level for this region. Three of five regions observed by {\nustar} on 2014 November 1 are well fit by homogeneous nanoflare models, while two regions with higher fluxes are not. Generally, the {\nustar} count spectra are well fit by nanoflare sequences with smaller heating amplitudes, shorter delays, and shorter durations than the allowed \textit{FOXSI-2} models. These apparent discrepancies are likely due to differences in spectral coverage between the two instruments and intrinsic differences among the regions. Steady heating ($t_N$ = $τ$) was ruled out with $>$99\% confidence for all regions observed by either instrument. △ Less

Submitted 8 August, 2018; originally announced August 2018.

Comments: 17 pages, 21 figures. Accepted for publication in The Astrophysical Journal

Efficient Calculation of Non-Local Thermodynamic Equilibrium Effects in Multithreaded Hydrodynamic Simulations of Solar Flares

Authors: Jeffrey W. Reep, Stephen J. Bradshaw, Nicholas A. Crump, Harry P. Warren

Abstract: Understanding the dynamics of the solar chromosphere is crucial to understanding energy transport across the solar atmosphere. The chromosphere is optically thick at many wavelengths and described by non-local thermodynamic equilibrium (NLTE), making it difficult to interpret observations. Furthermore, there is considerable evidence that the atmosphere is filamented, and that current instruments d… ▽ More Understanding the dynamics of the solar chromosphere is crucial to understanding energy transport across the solar atmosphere. The chromosphere is optically thick at many wavelengths and described by non-local thermodynamic equilibrium (NLTE), making it difficult to interpret observations. Furthermore, there is considerable evidence that the atmosphere is filamented, and that current instruments do not sufficiently resolve small scale features. In flares, it is likely that multithreaded models are required to describe the heating. The combination of NLTE effects and multithreaded modeling requires computationally demanding calculations, which has motivated the development of a model that can efficiently treat both. We describe the implementation of a solver in a hydrodynamic code for the hydrogen level populations that approximates the NLTE solutions. We derive an accurate electron density across the atmosphere, that includes the effects of non-equilibrium ionization for helium and metals. We show the effects on hydrodynamic simulations, which are used to synthesize light curves using a post-processing radiative transfer code. We demonstrate the utility of this model on IRIS observations of a small flare. We show that the Doppler shifts in Mg II, C II, and O I can be explained with a multithreaded model of loops subjected to electron beam heating, so long as NLTE effects are treated. The intensities, however, do not match observed values very well, which is due to assumptions about the initial atmosphere. We briefly show how altering the initial atmosphere can drastically alter line profiles, and therefore derived quantities, and suggest that it should be tuned to pre-flare observations. △ Less

Submitted 23 November, 2018; v1 submitted 25 June, 2018; originally announced June 2018.

Comments: Now accepted to ApJ. Comments and criticisms encouraged!

A Generative Model For Electron Paths

Authors: John Bradshaw, Matt J. Kusner, Brooks Paige, Marwin H. S. Segler, José Miguel Hernández-Lobato

Abstract: Chemical reactions can be described as the stepwise redistribution of electrons in molecules. As such, reactions are often depicted using `arrow-pushing' diagrams which show this movement as a sequence of arrows. We propose an electron path prediction model (ELECTRO) to learn these sequences directly from raw reaction data. Instead of predicting product molecules directly from reactant molecules i… ▽ More Chemical reactions can be described as the stepwise redistribution of electrons in molecules. As such, reactions are often depicted using `arrow-pushing' diagrams which show this movement as a sequence of arrows. We propose an electron path prediction model (ELECTRO) to learn these sequences directly from raw reaction data. Instead of predicting product molecules directly from reactant molecules in one shot, learning a model of electron movement has the benefits of (a) being easy for chemists to interpret, (b) incorporating constraints of chemistry, such as balanced atom counts before and after the reaction, and (c) naturally encoding the sparsity of chemical reactions, which usually involve changes in only a small number of atoms in the reactants.We design a method to extract approximate reaction paths from any dataset of atom-mapped reaction SMILES strings. Our model achieves excellent performance on an important subset of the USPTO reaction dataset, comparing favorably to the strongest baselines. Furthermore, we show that our model recovers a basic knowledge of chemistry without being explicitly trained to do so. △ Less

Submitted 20 March, 2019; v1 submitted 23 May, 2018; originally announced May 2018.

Are Generative Classifiers More Robust to Adversarial Attacks?

Authors: Yingzhen Li, John Bradshaw, Yash Sharma

Abstract: There is a rising interest in studying the robustness of deep neural network classifiers against adversaries, with both advanced attack and defence techniques being actively developed. However, most recent work focuses on discriminative classifiers, which only model the conditional distribution of the labels given the inputs. In this paper, we propose and investigate the deep Bayes classifier, whi… ▽ More There is a rising interest in studying the robustness of deep neural network classifiers against adversaries, with both advanced attack and defence techniques being actively developed. However, most recent work focuses on discriminative classifiers, which only model the conditional distribution of the labels given the inputs. In this paper, we propose and investigate the deep Bayes classifier, which improves classical naive Bayes with conditional deep generative models. We further develop detection methods for adversarial examples, which reject inputs with low likelihood under the generative model. Experimental results suggest that deep Bayes classifiers are more robust than deep discriminative classifiers, and that the proposed detection methods are effective against many recently proposed attacks. △ Less

Submitted 27 May, 2019; v1 submitted 19 February, 2018; originally announced February 2018.

Comments: ICML 2019

Adversarial Examples, Uncertainty, and Transfer Testing Robustness in Gaussian Process Hybrid Deep Networks

Authors: John Bradshaw, Alexander G. de G. Matthews, Zoubin Ghahramani

Abstract: Deep neural networks (DNNs) have excellent representative power and are state of the art classifiers on many tasks. However, they often do not capture their own uncertainties well making them less robust in the real world as they overconfidently extrapolate and do not notice domain shift. Gaussian processes (GPs) with RBF kernels on the other hand have better calibrated uncertainties and do not ov… ▽ More Deep neural networks (DNNs) have excellent representative power and are state of the art classifiers on many tasks. However, they often do not capture their own uncertainties well making them less robust in the real world as they overconfidently extrapolate and do not notice domain shift. Gaussian processes (GPs) with RBF kernels on the other hand have better calibrated uncertainties and do not overconfidently extrapolate far from data in their training set. However, GPs have poor representational power and do not perform as well as DNNs on complex domains. In this paper we show that GP hybrid deep networks, GPDNNs, (GPs on top of DNNs and trained end-to-end) inherit the nice properties of both GPs and DNNs and are much more robust to adversarial examples. When extrapolating to adversarial examples and testing in domain shift settings, GPDNNs frequently output high entropy class probabilities corresponding to essentially "don't know". GPDNNs are therefore promising as deep architectures that know when they don't know. △ Less

Submitted 8 July, 2017; originally announced July 2017.

The structure of Chariklo's rings from stellar occultations

Authors: D. Bérard, B. Sicardy, J. I. B. Camargo, J. Desmars, F. Braga-Ribas, J. L. Ortiz, R. Duffard, N. Morales, E. Meza, R. Leiva, G. Benedetti-Rossi, R. Vieira-Martins, A. R. Gomes-Júnior, M. Assafin, F. Colas, J. L. Dauvergne, P. Kervella, J. Lecacheux, L. Maquet, F. Vachier, S. Renner, B. Monard, A. A. Sickafoose, H. Breytenbach, A. Genade , et al. (57 additional authors not shown)

Abstract: Two narrow and dense rings (called C1R and C2R) were discovered around the Centaur object (10199) Chariklo during a stellar occultation observed on 2013 June 3. Following this discovery, we planned observations of several occultations by Chariklo's system in order to better characterize the physical properties of the ring and main body. Here, we use 12 successful occulations by Chariklo observed b… ▽ More Two narrow and dense rings (called C1R and C2R) were discovered around the Centaur object (10199) Chariklo during a stellar occultation observed on 2013 June 3. Following this discovery, we planned observations of several occultations by Chariklo's system in order to better characterize the physical properties of the ring and main body. Here, we use 12 successful occulations by Chariklo observed between 2014 and 2016. They provide ring profiles (physical width, opacity, edge structure) and constraints on the radii and pole position. Our new observations are currently consistent with the circular ring solution and pole position, to within the $\pm 3.3$ km formal uncertainty for the ring radii derived by Braga-Ribas et al. The six resolved C1R profiles reveal significant width variations from $\sim 5$ to 7.5 km. The width of the fainter ring C2R is less constrained, and may vary between 0.1 and 1 km. The inner and outer edges of C1R are consistent with infinitely sharp boundaries, with typical upper limits of one kilometer for the transition zone between the ring and empty space. No constraint on the sharpness of C2R's edges is available. A 1$σ$ upper limit of $\sim 20$ m is derived for the equivalent width of narrow (physical width <4 km) rings up to distances of 12,000 km, counted in the ring plane. △ Less

Submitted 18 September, 2017; v1 submitted 1 June, 2017; originally announced June 2017.

Journal ref: The Astronomical Journal, 154:144 (21pp), 2017 October

Study of the plutino object (208996) 2003 AZ84 from stellar occultations: size, shape and topographic features

Authors: A. Dias-Oliveira, B. Sicardy, J. L. Ortiz, F. Braga-Ribas, R. Leiva, R. Vieira-Martins, G. Benedetti-Rossi, J. I. B. Camargo, M. Assafin, A. R. Gomes-Junior, T. Baug, T. Chandrasekhar, J. Desmars, R. Duffard, P. Santos-Sanz, Z. Ergang, S. Ganesh, Y. Ikari, P. Irawati, J. Jain, Z. Liying, A. Richichi, Q. Shengbang, R. Behrend, Z. Benkhaldoun , et al. (38 additional authors not shown)

Abstract: We present results derived from four stellar occultations by the plutino object (208996) 2003~AZ$_{84}$, detected at January 8, 2011 (single-chord event), February 3, 2012 (multi-chord), December 2, 2013 (single-chord) and November 15, 2014 (multi-chord). Our observations rule out an oblate spheroid solution for 2003~AZ$_{84}$'s shape. Instead, assuming hydrostatic equilibrium, we find that a Jaco… ▽ More We present results derived from four stellar occultations by the plutino object (208996) 2003~AZ$_{84}$, detected at January 8, 2011 (single-chord event), February 3, 2012 (multi-chord), December 2, 2013 (single-chord) and November 15, 2014 (multi-chord). Our observations rule out an oblate spheroid solution for 2003~AZ$_{84}$'s shape. Instead, assuming hydrostatic equilibrium, we find that a Jacobi triaxial solution with semi axes $(470 \pm 20) \times (383 \pm 10) \times (245 \pm 8)$~km % axis ratios $b/a= 0.82 \pm 0.05$ and $c/a= 0.52 \pm 0.02$, can better account for all our occultation observations. Combining these dimensions with the rotation period of the body (6.75~h) and the amplitude of its rotation light curve, we derive a density $ρ=0.87 \pm 0.01$~g~cm$^{-3}$ a geometric albedo $p_V= 0.097 \pm 0.009$. A grazing chord observed during the 2014 occultation reveals a topographic feature along 2003~AZ$_{84}$'s limb, that can be interpreted as an abrupt chasm of width $\sim 23$~km and depth $> 8$~km or a smooth depression of width $\sim 80$~km and depth $\sim 13$~km (or an intermediate feature between those two extremes). △ Less

Submitted 30 May, 2017; originally announced May 2017.

Inference of Heating Properties from "Hot" Non-flaring Plasmas in Active Region Cores. II. Nanoflare Trains

Authors: W. T. Barnes, P. J. Cargill, S. J. Bradshaw

Abstract: Despite its prediction over two decades ago, the detection of faint, high-temperature ("hot") emission due to nanoflare heating in non-flaring active region cores has proved challenging. Using an efficient two-fluid hydrodynamic model, this paper investigates the properties of the emission expected from repeating nanoflares (a nanoflare train) of varying frequency as well as the separate heating o… ▽ More Despite its prediction over two decades ago, the detection of faint, high-temperature ("hot") emission due to nanoflare heating in non-flaring active region cores has proved challenging. Using an efficient two-fluid hydrodynamic model, this paper investigates the properties of the emission expected from repeating nanoflares (a nanoflare train) of varying frequency as well as the separate heating of electrons and ions. If the emission measure distribution ($\mathrm{EM}(T)$) peaks at $T = T_m$, we find that $\mathrm{EM}(T_m)$ is independent of details of the nanoflare train, and $\mathrm{EM}(T)$ above and below $T_m$ reflects different aspects of the heating. Below $T_m$ the main influence is the relationship of the waiting time between successive nanoflares to the nanoflare energy. Above $T_m$ power-law nanoflare distributions lead to an extensive plasma population not present in a monoenergetic train. Furthermore, in some cases characteristic features are present in $\mathrm{EM}(T)$. Such details may be detectable given adequate spectral resolution and a good knowledge of the relevant atomic physics. In the absence of such resolution we propose some metrics that can be used to infer the presence of "hot" plasma. △ Less

Submitted 19 October, 2016; originally announced October 2016.

Comments: 15 pages, 11 figures, accepted to ApJ

Inference of Heating Properties from "Hot" Non-flaring Plasmas in Active Region Cores I. Single Nanoflares

Authors: W. T. Barnes, P. J. Cargill, S. J. Bradshaw

Abstract: The properties expected of "hot" non-flaring plasmas due to nanoflare heating in active regions are investigated using hydrodynamic modeling tools, including a two-fluid development of the EBTEL code. Here we study a single nanoflare and show that while simple models predict an emission measure distribution extending well above 10 MK that is consistent with cooling by thermal conduction, many othe… ▽ More The properties expected of "hot" non-flaring plasmas due to nanoflare heating in active regions are investigated using hydrodynamic modeling tools, including a two-fluid development of the EBTEL code. Here we study a single nanoflare and show that while simple models predict an emission measure distribution extending well above 10 MK that is consistent with cooling by thermal conduction, many other effects are likely to limit the existence and detectability of such plasmas. These include: differential heating between electrons and ions, ionization non-equilibrium and, for short nanoflares, the time taken for the coronal density to increase. The most useful temperature range to look for this plasma, often called the "smoking gun" of nanoflare heating, lies between $10^{6.6}$ and $10^7$ K. Signatures of the actual heating may be detectable in some instances. △ Less

Submitted 16 August, 2016; originally announced August 2016.

Comments: Accepted for publication in ApJ, 16 pages, 8 figures

Chromospheric Nanoflares as a Source of Coronal Plasma: II. Repeating Nanoflares

Authors: Stephen J. Bradshaw, James A. Klimchuk

Abstract: The million degree plasma of the solar corona must be supplied by the underlying layers of the atmosphere. The mechanism and location of energy release, and the precise source of coronal plasma, remain unresolved. In earlier work we pursued the idea that warm plasma is supplied to the corona via direct heating of the chromosphere by nanoflares, contrary to the prevailing belief that the corona is… ▽ More The million degree plasma of the solar corona must be supplied by the underlying layers of the atmosphere. The mechanism and location of energy release, and the precise source of coronal plasma, remain unresolved. In earlier work we pursued the idea that warm plasma is supplied to the corona via direct heating of the chromosphere by nanoflares, contrary to the prevailing belief that the corona is heated in-situ and the chromosphere is subsequently energized and ablated by thermal conduction. We found that single (low-frequency) chromospheric nanoflares could not explain the observed intensities, Doppler-shifts, and red/blue asymmetries in Fe XII and XIV emission lines. In the present work we follow up on another suggestion that the corona could be powered by chromospheric nanoflares that repeat on a timescale substantially shorter than the cooling/draining timescale. That is, a single magnetic strand is re-supplied with coronal plasma before the existing plasma has time to cool and drain. We perform a series of hydrodynamic experiments and predict the Fe XII and XIV line intensities, Doppler-shifts, and red/blue asymmetries. We find that our predicted quantities disagree dramatically with observations and fully developed loop structures cannot be created by intermediate- or high-frequency chromospheric nanoflares. We conclude that the mechanism ultimately responsible for producing coronal plasma operates above the chromosphere, but this does not preclude the possibility of a similar mechanism powering the chromosphere; extreme examples of which may be responsible for heating chromospheric plasma to transition region temperatures (e.g. type II spicules). △ Less

Submitted 22 March, 2016; originally announced March 2016.

Journal ref: ApJ, 811, 129 (2015)