Showing 1–45 of 45 results for author: Leung, A

UICoder: Finetuning Large Language Models to Generate User Interface Code through Automated Feedback

Authors: Jason Wu, Eldon Schoop, Alan Leung, Titus Barik, Jeffrey P. Bigham, Jeffrey Nichols

Abstract: Large language models (LLMs) struggle to consistently generate UI code that compiles and produces visually relevant designs. Existing approaches to improve generation rely on expensive human feedback or distilling a proprietary model. In this paper, we explore the use of automated feedback (compilers and multi-modal models) to guide LLMs to generate high-quality UI code. Our method starts with an… ▽ More Large language models (LLMs) struggle to consistently generate UI code that compiles and produces visually relevant designs. Existing approaches to improve generation rely on expensive human feedback or distilling a proprietary model. In this paper, we explore the use of automated feedback (compilers and multi-modal models) to guide LLMs to generate high-quality UI code. Our method starts with an existing LLM and iteratively produces improved models by self-generating a large synthetic dataset using an original model, applying automated tools to aggressively filter, score, and de-duplicate the data into a refined higher quality dataset. The original LLM is improved by finetuning on this refined dataset. We applied our approach to several open-source LLMs and compared the resulting performance to baseline models with both automated metrics and human preferences. Our evaluation shows the resulting models outperform all other downloadable baselines and approach the performance of larger proprietary models. △ Less

Submitted 11 June, 2024; originally announced June 2024.

Comments: Accepted to NAACL 2024

Language Models are Alignable Decision-Makers: Dataset and Application to the Medical Triage Domain

Authors: Brian Hu, Bill Ray, Alice Leung, Amy Summerville, David Joy, Christopher Funk, Arslan Basharat

Abstract: In difficult decision-making scenarios, it is common to have conflicting opinions among expert human decision-makers as there may not be a single right answer. Such decisions may be guided by different attributes that can be used to characterize an individual's decision. We introduce a novel dataset for medical triage decision-making, labeled with a set of decision-maker attributes (DMAs). This da… ▽ More In difficult decision-making scenarios, it is common to have conflicting opinions among expert human decision-makers as there may not be a single right answer. Such decisions may be guided by different attributes that can be used to characterize an individual's decision. We introduce a novel dataset for medical triage decision-making, labeled with a set of decision-maker attributes (DMAs). This dataset consists of 62 scenarios, covering six different DMAs, including ethical principles such as fairness and moral desert. We present a novel software framework for human-aligned decision-making by utilizing these DMAs, paving the way for trustworthy AI with better guardrails. Specifically, we demonstrate how large language models (LLMs) can serve as ethical decision-makers, and how their decisions can be aligned to different DMAs using zero-shot prompting. Our experiments focus on different open-source models with varying sizes and training techniques, such as Falcon, Mistral, and Llama 2. Finally, we also introduce a new form of weighted self-consistency that improves the overall quantified performance. Our results provide new research directions in the use of LLMs as alignable decision-makers. The dataset and open-source software are publicly available at: https://github.com/ITM-Kitware/llm-alignable-dm. △ Less

Submitted 10 June, 2024; originally announced June 2024.

Comments: 15 pages total (including appendix), NAACL 2024 Industry Track

Analytical Correlation in the H$_{2}$ Molecule from the Independent Atom Ansatz

Authors: Alanna 'Lanie' Leung, Alexander V. Mironenko

Abstract: The independent atom ansatz of density functional theory yields an accurate analytical expression for dynamic correlation energy in the H$_{2}$ molecule: $E_{c} = 0.5(1 - \sqrt{2})(ab|ba)$ for the atom-additive self-consistent density $ρ= |a|^{2} + |b|^{2}$. Combined with exact atomic self-exchange, it recovers more than 99.5 % of nearly exact SCAN exchange-correlation energy at R > 0.5 $Å$, diffe… ▽ More The independent atom ansatz of density functional theory yields an accurate analytical expression for dynamic correlation energy in the H$_{2}$ molecule: $E_{c} = 0.5(1 - \sqrt{2})(ab|ba)$ for the atom-additive self-consistent density $ρ= |a|^{2} + |b|^{2}$. Combined with exact atomic self-exchange, it recovers more than 99.5 % of nearly exact SCAN exchange-correlation energy at R > 0.5 $Å$, differing by less than 0.12 eV. The total energy functional correctly dissociates the H-H bond and yields absolute errors of 0.002 $Å$, 0.19 eV, and 13 cm$^{-1}$ relative to experiment at the tight binding computational cost. The chemical bond formation is attributed to the asymptotic Heitler-London resonance of quasi-orthogonal atomic states ($- (ab|ba)$) with no contributions from kinetic energy or charge accumulation in the bond. △ Less

Submitted 20 May, 2024; originally announced May 2024.

Comments: 7 pages, 4 figures

BISCUIT: Scaffolding LLM-Generated Code with Ephemeral UIs in Computational Notebooks

Authors: Ruijia Cheng, Titus Barik, Alan Leung, Fred Hohman, Jeffrey Nichols

Abstract: Novices frequently engage with machine learning tutorials in computational notebooks and have been adopting code generation technologies based on large language models (LLMs). However, they encounter difficulties in understanding and working with code produced by LLMs. To mitigate these challenges, we introduce a novel workflow into computational notebooks that augments LLM-based code generation w… ▽ More Novices frequently engage with machine learning tutorials in computational notebooks and have been adopting code generation technologies based on large language models (LLMs). However, they encounter difficulties in understanding and working with code produced by LLMs. To mitigate these challenges, we introduce a novel workflow into computational notebooks that augments LLM-based code generation with an additional ephemeral UI step, offering users UI-based scaffolds as an intermediate stage between user prompts and code generation. We present this workflow in BISCUIT, an extension for JupyterLab that provides users with ephemeral UIs generated by LLMs based on the context of their code and intentions, scaffolding users to understand, guide, and explore with LLM-generated code. Through a user study where 10 novices used BISCUIT for machine learning tutorials, we discover that BISCUIT offers user semantic representation of code to aid their understanding, reduces the complexity of prompt engineering, and creates a playground for users to explore different variables and iterate on their ideas. We discuss the implications of our findings for UI-centric interactive paradigm in code generation LLMs. △ Less

Submitted 12 April, 2024; v1 submitted 10 April, 2024; originally announced April 2024.

Autoregressive Search of Gravitational Waves: Denoising

Authors: Sangin Kim, C. Y. Hui, Jianqi Yan, Alex P. Leung, Kwangmin Oh, A. K. H. Kong, L. C. -C. Lin, Kwan-Lok Li

Abstract: Because of the small strain amplitudes of gravitational-wave (GW) signals, unveiling them in the presence of detector/environmental noise is challenging. For visualizing the signals and extracting its waveform for a comparison with theoretical prediction, a frequency-domain whitening process is commonly adopted for filtering the data. In this work, we propose an alternative template-free framework… ▽ More Because of the small strain amplitudes of gravitational-wave (GW) signals, unveiling them in the presence of detector/environmental noise is challenging. For visualizing the signals and extracting its waveform for a comparison with theoretical prediction, a frequency-domain whitening process is commonly adopted for filtering the data. In this work, we propose an alternative template-free framework based on autoregressive modeling for denoising the GW data and extracting the waveform. We have tested our framework on extracting the injected signals from the simulated data as well as a series of known compact binary coalescence (CBC) events from the LIGO data. Comparing with the conventional whitening procedure, our methodology generally yields improved cross-correlation and reduced root mean square errors with respect to the signal model. △ Less

Submitted 8 April, 2024; originally announced April 2024.

Comments: Phys. Rev. D in press, 16 pages, 11 figures, 1 table

Graphs with queue number three and unbounded stack number

Authors: Yui Hin Arvin Leung

Abstract: We prove that the graphs $T\boxslash P$ have unbounded stack number and queue number $3$, where $T$ is a tree and $P$ is a path, and $\boxslash$ denotes the graph strong product but with one of the directions removed. The previous best known results is that graphs with queue number $4$ can have unbounded stack number. We prove that the graphs $T\boxslash P$ have unbounded stack number and queue number $3$, where $T$ is a tree and $P$ is a path, and $\boxslash$ denotes the graph strong product but with one of the directions removed. The previous best known results is that graphs with queue number $4$ can have unbounded stack number. △ Less

Submitted 27 March, 2023; originally announced March 2023.

Comments: 31 pages, 7 figures

Multi-scale Hybridized Topic Modeling: A Pipeline for Analyzing Unstructured Text Datasets via Topic Modeling

Authors: Keyi Cheng, Stefan Inzer, Adrian Leung, Xiaoxian Shen, Michael Perlmutter, Michael Lindstrom, Joyce Chew, Todd Presner, Deanna Needell

Abstract: We propose a multi-scale hybridized topic modeling method to find hidden topics from transcribed interviews more accurately and efficiently than traditional topic modeling methods. Our multi-scale hybridized topic modeling method (MSHTM) approaches data at different scales and performs topic modeling in a hierarchical way utilizing first a classical method, Nonnegative Matrix Factorization, and th… ▽ More We propose a multi-scale hybridized topic modeling method to find hidden topics from transcribed interviews more accurately and efficiently than traditional topic modeling methods. Our multi-scale hybridized topic modeling method (MSHTM) approaches data at different scales and performs topic modeling in a hierarchical way utilizing first a classical method, Nonnegative Matrix Factorization, and then a transformer-based method, BERTopic. It harnesses the strengths of both NMF and BERTopic. Our method can help researchers and the public better extract and interpret the interview information. Additionally, it provides insights for new indexing systems based on the topic level. We then deploy our method on real-world interview transcripts and find promising results. △ Less

Submitted 24 November, 2022; originally announced November 2022.

On Improving the Performance of Glitch Classification for Gravitational Wave Detection by using Generative Adversarial Networks

Authors: Jianqi Yan, Alex P. Leung, David C. Y. Hui

Abstract: Spectrogram classification plays an important role in analyzing gravitational wave data. In this paper, we propose a framework to improve the classification performance by using Generative Adversarial Networks (GANs). As substantial efforts and expertise are required to annotate spectrograms, the number of training examples is very limited. However, it is well known that deep networks can perform… ▽ More Spectrogram classification plays an important role in analyzing gravitational wave data. In this paper, we propose a framework to improve the classification performance by using Generative Adversarial Networks (GANs). As substantial efforts and expertise are required to annotate spectrograms, the number of training examples is very limited. However, it is well known that deep networks can perform well only when the sample size of the training set is sufficiently large. Furthermore, the imbalanced sample sizes in different classes can also hamper the performance. In order to tackle these problems, we propose a GAN-based data augmentation framework. While standard data augmentation methods for conventional images cannot be applied on spectrograms, we found that a variant of GANs, ProGAN, is capable of generating high-resolution spectrograms which are consistent with the quality of the high-resolution original images and provide a desirable diversity. We have validated our framework by classifying glitches in the {\it Gravity Spy} dataset with the GAN-generated spectrograms for training. We show that the proposed method can provide an alternative to transfer learning for the classification of spectrograms using deep networks, i.e. using a high-resolution GAN for data augmentation instead. Furthermore, fluctuations in classification performance with small sample sizes for training and evaluation can be greatly reduced. Using the trained network in our framework, we have also examined the spectrograms with label anomalies in {\it Gravity Spy}. △ Less

Submitted 8 July, 2022; originally announced July 2022.

Comments: Accepted for publication in MNRAS, 16 pages, 14 figures, 5 tables

Observation of cross phase modulation in cold atom gradient echo memory

Authors: Anthony C. Leung, K. S. Ida Melody, Aaron D. Tranter, Karun V. Paul, Geoff T. Campbell, ** Koy Lam, Ben C. Buchler

Abstract: Strong nonlinear interactions between single photons have important applications in optical quantum information processing. Demonstrations of these interactions in cold atomic ensembles have largely been limited to exploiting slow light generated using electromagnetically induced transparency (EIT). However, these EIT implementations have limited achievable phase shifts due to spontaneous emission… ▽ More Strong nonlinear interactions between single photons have important applications in optical quantum information processing. Demonstrations of these interactions in cold atomic ensembles have largely been limited to exploiting slow light generated using electromagnetically induced transparency (EIT). However, these EIT implementations have limited achievable phase shifts due to spontaneous emission. Here, we demonstrate and characterize a scheme free from these limitations using gradient echo memory with inferred single photon phase shifts of $0.07\pm0.02$ $μ\text{rad}$. Excellent agreement with theoretical modelling was observed. Degradation of memory efficiency was observed for large phase shifts but strategies to overcome that are presented. △ Less

Submitted 20 May, 2022; originally announced May 2022.

Comments: 7 pages, 6 figures

Efficient, ever-ready quantum memory at room temperature for single photons

Authors: Anthony C. Leung, W. Y. Sarah Lau, Aaron D. Tranter, Karun V. Paul, Markus Rambach, Ben C. Buchler, ** Koy Lam, Andrew G. White, Till J. Weinhold

Abstract: Efficient quantum memories will be an essential building block of large scale networked quantum systems and provide a link between flying photonic qubits and atomic or quasi-atomic local quantum processors. To provide a path to scalability avoidance of bulky, difficult to maintain systems such as high vacuum and low temperature cryogenics is imperative. Memory efficiencies above 50% are required t… ▽ More Efficient quantum memories will be an essential building block of large scale networked quantum systems and provide a link between flying photonic qubits and atomic or quasi-atomic local quantum processors. To provide a path to scalability avoidance of bulky, difficult to maintain systems such as high vacuum and low temperature cryogenics is imperative. Memory efficiencies above 50% are required to be operating above the quantum no-cloning limit. Such high efficiencies have only been achieved in systems with photon sources tailored to the memory bandwidth. In this paper we explore the combination of an ultralow spectral bandwidth source of single photons from cavity-enhanced spontaneous parametric down-conversion with a gas-ensemble atomic memory. Our rubidium vapour gradient echo memory achieves 84$\pm$3% recall efficiency of single photons: a record for an always-ready, hot, and vacuum system free optical memory. △ Less

Submitted 29 March, 2022; v1 submitted 22 March, 2022; originally announced March 2022.

Comments: 11 pages, 10 figures

Two-center basis generator method calculations for Li$^{3+}$, C$^{3+}$ and O$^{3+}$ ion impact on ground state hydrogen

Authors: Anthony C. K. Leung, Tom Kirchner

Abstract: The two-center basis generator method is used to obtain cross sections for excitation, capture, and ionization in Li$^{3+}$, C$^{3+}$, and O$^{3+}$ collisions with ground-state hydrogen at projectile energies from 1 to 100 keV/u. The interaction of the C$^{3+}$ and O$^{3+}$ projectiles with the active electron is represented by a model potential. Comparisons of cross sections with previously repor… ▽ More The two-center basis generator method is used to obtain cross sections for excitation, capture, and ionization in Li$^{3+}$, C$^{3+}$, and O$^{3+}$ collisions with ground-state hydrogen at projectile energies from 1 to 100 keV/u. The interaction of the C$^{3+}$ and O$^{3+}$ projectiles with the active electron is represented by a model potential. Comparisons of cross sections with previously reported data show overall good agreement while discrepancies in capture for C$^{3+}$ collisions at low energies are noted. The present results show that excitation and ionization are similar across the three collision systems, which indicates that these cross sections are mostly dependent on the net charge of the projectile only. The situation is different for the capture channel. △ Less

Submitted 21 January, 2022; v1 submitted 23 December, 2021; originally announced December 2021.

3D modelling of survey scene from images enhanced with a multi-exposure fusion

Authors: Kwok-Leung Chan, Li** Li, Arthur Wing-Tak Leung, Ho-Yin Chan

Abstract: In current practice, scene survey is carried out by workers using total stations. The method has high accuracy, but it incurs high costs if continuous monitoring is needed. Techniques based on photogrammetry, with the relatively cheaper digital cameras, have gained wide applications in many fields. Besides point measurement, photogrammetry can also create a three-dimensional (3D) model of the scen… ▽ More In current practice, scene survey is carried out by workers using total stations. The method has high accuracy, but it incurs high costs if continuous monitoring is needed. Techniques based on photogrammetry, with the relatively cheaper digital cameras, have gained wide applications in many fields. Besides point measurement, photogrammetry can also create a three-dimensional (3D) model of the scene. Accurate 3D model reconstruction depends on high quality images. Degraded images will result in large errors in the reconstructed 3D model. In this paper, we propose a method that can be used to improve the visibility of the images, and eventually reduce the errors of the 3D scene model. The idea is inspired by image dehazing. Each original image is first transformed into multiple exposure images by means of gamma-correction operations and adaptive histogram equalization. The transformed images are analyzed by the computation of the local binary patterns. The image is then enhanced, with each pixel generated from the set of transformed image pixels weighted by a function of the local pattern feature and image saturation. Performance evaluation has been performed on benchmark image dehazing datasets. Experimentations have been carried out on outdoor and indoor surveys. Our analysis finds that the method works on different types of degradation that exist in both outdoor and indoor images. When fed into the photogrammetry software, the enhanced images can reconstruct 3D scene models with sub-millimeter mean errors. △ Less

Submitted 10 November, 2021; originally announced November 2021.

Contact Graph Epidemic Modelling of COVID-19 for Transmission and Intervention Strategies

Authors: Abby Leung, Xiaoye Ding, Shenyang Huang, Reihaneh Rabbany

Abstract: The coronavirus disease 2019 (COVID-19) pandemic has quickly become a global public health crisis unseen in recent years. It is known that the structure of the human contact network plays an important role in the spread of transmissible diseases. In this work, we study a structure aware model of COVID-19 CGEM. This model becomes similar to the classical compartment-based models in epidemiology if… ▽ More The coronavirus disease 2019 (COVID-19) pandemic has quickly become a global public health crisis unseen in recent years. It is known that the structure of the human contact network plays an important role in the spread of transmissible diseases. In this work, we study a structure aware model of COVID-19 CGEM. This model becomes similar to the classical compartment-based models in epidemiology if we assume the contact network is a Erdos-Renyi (ER) graph, i.e. everyone comes into contact with everyone else with the same probability. In contrast, CGEM is more expressive and allows for plugging in the actual contact networks, or more realistic proxies for it. Moreover, CGEM enables more precise modelling of enforcing and releasing different non-pharmaceutical intervention (NPI) strategies. Through a set of extensive experiments, we demonstrate significant differences between the epidemic curves when assuming different underlying structures. More specifically we demonstrate that the compartment-based models are overestimating the spread of the infection by a factor of 3, and under some realistic assumptions on the compliance factor, underestimating the effectiveness of some of NPIs, mischaracterizing others (e.g. predicting a later peak), and underestimating the scale of the second peak after reopening. △ Less

Submitted 6 October, 2020; originally announced October 2020.

Floquet spin states in OLEDs

Authors: S. Jamali, V. V. Mkhitaryan, H. Malissa, A. Nahlawi, H. Popli, T. Grünbaum, S. Bange, S. Milster, D. Stoltzfus, A. E. Leung, T. A. Darwish, P. L. Burn, J. M. Lupton, C. Boehme

Abstract: Weakly spin-orbit coupled electron and hole spins in organic light-emitting diodes (OLEDs) constitute near-perfect two-level systems to explore the interaction of light and matter in the ultrastrong-drive regime. Under such highly non-perturbative conditions, the frequency at which the spin oscillates between states, the Rabi frequency, becomes comparable to its natural resonance frequency, the La… ▽ More Weakly spin-orbit coupled electron and hole spins in organic light-emitting diodes (OLEDs) constitute near-perfect two-level systems to explore the interaction of light and matter in the ultrastrong-drive regime. Under such highly non-perturbative conditions, the frequency at which the spin oscillates between states, the Rabi frequency, becomes comparable to its natural resonance frequency, the Larmor frequency. For such conditions, we develop an intuitive understanding of the emergence of hybrid light-matter states, illustrating how dipole-forbidden multiple-quantum transitions at integer and fractional g-factors arise. A rigorous theoretical treatment of the phenomena comes from a Floquet-style solution to the time-dependent Hamiltonian of the electron-hole spin pair under resonant drive. To probe these phenomena experimentally requires both the development of a magnetic-resonance setup capable of supporting oscillating driving fields comparable in magnitude to the static field defining the Zeeman splitting; and an organic semiconductor which is characterized by minimal inhomogeneous broadening so as to allow the non-linear light-matter interactions to be resolved. The predicted exotic resonance features associated with the Floquet states are indeed found experimentally in measurements of spin-dependent steady-state OLED current under resonant drive, demonstrating that complex hybrid light-matter spin excitations can be formed and probed at room temperature. The spin-Dicke state arising under strong drive is insensitive to power broadening so that the Bloch-Siegert shift of the resonance becomes apparent, implying long coherence times of the dressed spin state with potential applicability for quantum sensing. △ Less

Submitted 5 October, 2020; originally announced October 2020.

Comments: Manuscript: 26 pages, 3 figures; supplementary information: 28 pages, 7 figures

Incorporating Dynamic Flight Network in SEIR to Model Mobility between Populations

Authors: Xiaoye Ding, Shenyang Huang, Abby Leung, Reihaneh Rabbany

Abstract: Current efforts of modelling COVID-19 are often based on the standard compartmental models such as SEIR and their variations. As pre-symptomatic and asymptomatic cases can spread the disease between populations through travel, it is important to incorporate mobility between populations into the epidemiological modelling. In this work, we propose to modify the commonly-used SEIR model to account fo… ▽ More Current efforts of modelling COVID-19 are often based on the standard compartmental models such as SEIR and their variations. As pre-symptomatic and asymptomatic cases can spread the disease between populations through travel, it is important to incorporate mobility between populations into the epidemiological modelling. In this work, we propose to modify the commonly-used SEIR model to account for the dynamic flight network, by estimating the imported cases based on the air traffic volume as well as the test positive rate at the source. This modification, called Flight-SEIR, can potentially enable 1). early detection of outbreaks due to imported pre-symptomatic and asymptomatic cases, 2). more accurate estimation of the reproduction number and 3). evaluation of the impact of travel restrictions and the implications of lifting these measures. The proposed Flight-SEIR is essential in navigating through this pandemic and the next ones, given how interconnected our world has become. △ Less

Submitted 3 October, 2020; originally announced October 2020.

Correlative Synchrotron X-ray Imaging and Diffraction of Directed Energy Deposition Additive Manufacturing

Authors: Yunhui Chen, Samuel J. Clark, David M. Collins, Sebastian Marussi, Simon A. Hunt, Danielle M. Fenech, Thomas Connolley, Robert C. Atwood, Oxana V. Magdysyuk, Gavin J. Baxter, Martyn A. Jones, Chu Lun Alex Leung, Peter D. Lee

Abstract: The governing mechanistic behaviour of Directed Energy Deposition Additive Manufacturing (DED-AM) is revealed by a combined in situ and operando synchrotron X-ray imaging and diffraction study of a nickel-base superalloy, IN718. Using a unique process replicator, real-space phase-contrast imaging enables quantification of the melt-pool boundary and flow dynamics during solidification. This imaging… ▽ More The governing mechanistic behaviour of Directed Energy Deposition Additive Manufacturing (DED-AM) is revealed by a combined in situ and operando synchrotron X-ray imaging and diffraction study of a nickel-base superalloy, IN718. Using a unique process replicator, real-space phase-contrast imaging enables quantification of the melt-pool boundary and flow dynamics during solidification. This imaging knowledge informed precise diffraction measurements of temporally resolved microstructural phases during transformation and stress development with a spatial resolution of 100 $μ$m. The diffraction quantified thermal gradient enabled a dendritic solidification microstructure to be predicted and coupled to the stress orientation and magnitude. The fast cooling rate entirely suppressed the formation of secondary phases or recrystallisation in the solid-state. Upon solidification, the stresses rapidly increase to the yield strength during cooling. This insight, combined with IN718 $'$s large solidification range suggests that the accumulated plasticity exhausts the alloy$'$s ductility, causing liquation cracking. This study has revealed additional fundamental mechanisms governing the formation of highly non-equilibrium microstructures during DED-AM. △ Less

Submitted 16 September, 2020; originally announced September 2020.

DRIFT: Deep Reinforcement Learning for Functional Software Testing

Authors: Luke Harries, Rebekah Storan Clarke, Timothy Chapman, Swamy V. P. L. N. Nallamalli, Levent Ozgur, Shuktika Jain, Alex Leung, Steve Lim, Aaron Dietrich, José Miguel Hernández-Lobato, Tom Ellis, Cheng Zhang, Kamil Ciosek

Abstract: Efficient software testing is essential for productive software development and reliable user experiences. As human testing is inefficient and expensive, automated software testing is needed. In this work, we propose a Reinforcement Learning (RL) framework for functional software testing named DRIFT. DRIFT operates on the symbolic representation of the user interface. It uses Q-learning through Ba… ▽ More Efficient software testing is essential for productive software development and reliable user experiences. As human testing is inefficient and expensive, automated software testing is needed. In this work, we propose a Reinforcement Learning (RL) framework for functional software testing named DRIFT. DRIFT operates on the symbolic representation of the user interface. It uses Q-learning through Batch-RL and models the state-action value function with a Graph Neural Network. We apply DRIFT to testing the Windows 10 operating system and show that DRIFT can robustly trigger the desired software functionality in a fully automated manner. Our experiments test the ability to perform single and combined tasks across different applications, demonstrating that our framework can efficiently test software with a large range of testing objectives. △ Less

Submitted 16 July, 2020; originally announced July 2020.

In situ and Operando X-ray Imaging of Directed Energy Deposition Additive Manufacturing

Authors: Yunhui Chen, Samuel J. Clark, Lorna Sinclair, Chu Lun Alex Leung, Sebastian Marussi, Thomas Connolley, Oxana V. Magdysyuk, Robert C. Atwood, Gavin J. Baxter, Martyn A. Jones, David G. McCartney, Iain Todd, Peter D. Lee

Abstract: The mechanical performance of Directed Energy Deposition Additive Manufactured (DED-AM) components can be highly material dependent. Through in situ and operando synchrotron X-ray imaging we capture the underlying phenomena controlling build quality of stainless steel (SS316) and titanium alloy (Ti6242 or Ti-6Al-2Sn-4Zr-2Mo). We reveal three mechanisms influencing the build efficiency of titanium… ▽ More The mechanical performance of Directed Energy Deposition Additive Manufactured (DED-AM) components can be highly material dependent. Through in situ and operando synchrotron X-ray imaging we capture the underlying phenomena controlling build quality of stainless steel (SS316) and titanium alloy (Ti6242 or Ti-6Al-2Sn-4Zr-2Mo). We reveal three mechanisms influencing the build efficiency of titanium alloys compared to stainless steel: blown powder sintering; reduced melt-pool wetting due to the sinter; and pore pushing in the melt-pool. The former two directly increase lack of fusion porosity, while the later causes end of track porosity. Each phenomenon influences the melt-pool characteristics, wetting of the substrate and hence build efficacy and undesirable microstructural feature formation. We demonstrate that porosity is related to powder characteristics, pool flow, and solidification front morphology. Our results clarify DED-AM process dynamics, illustrating why each alloy builds differently, facilitating the wider application of additive manufacturing to new materials. △ Less

Submitted 16 June, 2020; originally announced June 2020.

Comments: 20 pages, 4 figures, submitted to Science Advances

Searches for Pulsar-like Candidates from Unidentified Objects in the Third Catalog of Hard Fermi-LAT (3FHL) sources with Machine Learning Techniques

Authors: C. Y. Hui, Jongsu Lee, K. L. Li, Sangin Kim, Kwangmin Oh, Shengda Luo, Alex P. Leung, A. K. H. Kong, J. Takata, K. S. Cheng

Abstract: We report the results of searching pulsar-like candidates from the unidentified objects in the $3^{\rm rd}$ Catalog of Hard Fermi-LAT sources (3FHL). Using a machine-learning based classification scheme with a nominal accuracy of $\sim98\%$, we have selected 27 pulsar-like objects from 200 unidentified 3FHL sources for an identification campaign. Using archival data, X-ray sources are found within… ▽ More We report the results of searching pulsar-like candidates from the unidentified objects in the $3^{\rm rd}$ Catalog of Hard Fermi-LAT sources (3FHL). Using a machine-learning based classification scheme with a nominal accuracy of $\sim98\%$, we have selected 27 pulsar-like objects from 200 unidentified 3FHL sources for an identification campaign. Using archival data, X-ray sources are found within the $γ-$ray error ellipses of 10 3FHL pulsar-like candidates. Within the error circles of the much better constrained X-ray positions, we have also searched for the optical/infrared counterparts and examined their spectral energy distributions. Among our short-listed candidates, the most secure identification is the association of 3FHL J1823.3-1339 and its X-ray counterpart with the globular cluster Mercer 5. The $γ-$rays from the source can be contributed by a population of millisecond pulsars residing in the cluster. This makes Mercer 5 as one of the slowly growing hard $γ-$ray population of globular clusters with emission $>10$ GeV. Very recently, another candidate picked by our classification scheme, 3FHL J1405.1-6118, has been identified as a new $γ-$ray binary with an orbital period of $13.7$ days. Our X-ray analysis with a short Chandra observation has found a possible periodic signal candidate of $\sim1.4$ hrs and a putative extended X-ray tail of $\sim20$ arcsec long. Spectral energy distribution of its optical/infrared counterpart conforms with a blackbody of $T_{\rm bb}\sim40000$ K and $R_{\rm bb}\sim12R_{\odot}$ at a distance of 7.7 kpc. This is consistent with its identification as an early O star as found by infrared spectroscopy. △ Less

Submitted 30 April, 2020; v1 submitted 22 April, 2020; originally announced April 2020.

Comments: MNRAS in press, 19 pages, 13 figures, 8 tables

An investigation on the factors affecting machine learning classifications in $γ$-ray astronomy

Authors: Shengda Luo, Alex P. Leung, C. Y. Hui, K. L. Li

Abstract: We have investigated a number of factors that can have significant impacts on the classification performance of $γ$-ray sources detected by Fermi Large Area Telescope (LAT) with machine learning techniques. We show that a framework of automatic feature selection can construct a simple model with a small set of features which yields better performance over previous results. Secondly, because of the… ▽ More We have investigated a number of factors that can have significant impacts on the classification performance of $γ$-ray sources detected by Fermi Large Area Telescope (LAT) with machine learning techniques. We show that a framework of automatic feature selection can construct a simple model with a small set of features which yields better performance over previous results. Secondly, because of the small sample size of the training/test sets of certain classes in $γ$-ray, nested re-sampling and cross-validations are suggested for quantifying the statistical fluctuations of the quoted accuracy. We have also constructed a test set by cross-matching the identified active galactic nuclei (AGNs) and the pulsars (PSRs) in the Fermi LAT eight-year point source catalog (4FGL) with those unidentified sources in the previous 3$^{\rm rd}$ Fermi LAT Source Catalog (3FGL). Using this cross-matched set, we show that some features used for building classification model with the identified source can suffer from the problem of covariate shift, which can be a result of various observational effects. This can possibly hamper the actual performance when one applies such model in classifying unidentified sources. Using our framework, both AGN/PSR and young pulsar (YNG)/millisecond pulsar (MSP) classifiers are automatically updated with the new features and the enlarged training samples in 4FGL catalog incorporated. Using a two-layer model with these updated classifiers, we have selected 20 promising MSP candidates with confidence scores $>98\%$ from the unidentified sources in 4FGL catalog which can provide inputs for a multi-wavelength identification campaign. △ Less

Submitted 15 January, 2020; v1 submitted 13 January, 2020; originally announced January 2020.

Comments: 15 pages, 10 figures, 11 tables, accepted for publication in MNRAS

Perdeuteration of poly[2-methoxy-5-(2'-ethylhexyloxy)-1,4-phenylenevinylene] (d-MEHPPV): control of microscopic charge-carrier spin-spin coupling and of magnetic-field effects in optoelectronic devices

Authors: Dani M. Stoltzfus, Gajadhar Joshi, Henna Popli, Shirin Jamali, Marzieh Kavand, Sebastian Milster, Tobias Grünbaum, Sebastian Bange, Adnan Nahlawi, Mandefro Y. Teferi, Sabastian I. Atwood, Anna E. Leung, Tamim A. Darwish, Hans Malissa, Paul L. Burn, John M. Lupton, Christoph Boehme

Abstract: Control of the effective local hyperfine fields in a conjugated polymer, poly[2-methoxy-5-(2'-ethylhexyloxy)-1,4-phenylenevinylene] (MEHPPV), by isotopic engineering is reported. These fields, evident as a frequency-independent line broadening mechanism in electrically detected magnetic resonance spectroscopy (EDMR), originate from the unresolved hyperfine coupling between the electronic spin of c… ▽ More Control of the effective local hyperfine fields in a conjugated polymer, poly[2-methoxy-5-(2'-ethylhexyloxy)-1,4-phenylenevinylene] (MEHPPV), by isotopic engineering is reported. These fields, evident as a frequency-independent line broadening mechanism in electrically detected magnetic resonance spectroscopy (EDMR), originate from the unresolved hyperfine coupling between the electronic spin of charge carrier pairs and the nuclear spins of surrounding hydrogen isotopes. The room temperature study of effects caused by complete deuteration of this polymer through magnetoresistance, magnetoelectroluminescence, coherent pulsed and multi-frequency EDMR, as well as inverse spin-Hall effect measurements, confirm the weak hyperfine broadening of charge carrier magnetic resonance lines. As a consequence, we can resolve coherent charge-carrier spin-beating, allowing for direct measurements of the magnitude of electronic spin-spin interactions. In addition, the weak hyperfine coupling allows us to resolve substantial spin-orbit coupling effects in EDMR spectra, even at low magnetic field strengths. These results illustrate the dramatic influence of hyperfine fields on the spin physics of organic light-emitting diode (OLED) materials at room temperature, and point to routes to reaching exotic ultra-strong resonant-drive regimes needed for the study of light-matter interactions. △ Less

Submitted 26 September, 2019; originally announced September 2019.

Comments: 22 pages (8 pages manuscript, 14 pages supplementary information)

Proton impact on ground and excited states of atomic hydrogen

Authors: Anthony C. K. Leung, Tom Kirchner

Abstract: The processes of electron excitation, capture, and ionization were investigated in proton collisions with atomic hydrogen in the initial $n=1$ and $n=2$ states at impact energies from 1 to 300 keV. The theoretical analysis is based on the close-coupling two-center basis generator method in the semiclassical approximation. Calculated cross sections are compared with previous results which include d… ▽ More The processes of electron excitation, capture, and ionization were investigated in proton collisions with atomic hydrogen in the initial $n=1$ and $n=2$ states at impact energies from 1 to 300 keV. The theoretical analysis is based on the close-coupling two-center basis generator method in the semiclassical approximation. Calculated cross sections are compared with previous results which include data obtained from classical-trajectory Monte Carlo, convergent close-coupling, and other two-center atomic orbital expansion approaches. There is an overall good agreement in the capture and excitation cross sections while there are some discrepancies in the ionization results at certain impact energies. These discrepancies in the present results can be partially understood through the use of a $1/n^{3}$ scaling model. △ Less

Submitted 6 December, 2019; v1 submitted 18 July, 2019; originally announced July 2019.

Comments: 7 pages, 4 figures, Eur. Phys. J. D. Final Accepted Version

Journal ref: Eur. Phys. J. D (2019) 73: 246

General anesthesia reduces complexity and temporal asymmetry of the informational structures derived from neural recordings in Drosophila

Authors: Roberto N. Muñoz, Angus Leung, Aidan Zecevik, Felix A. Pollock, Dror Cohen, Bruno van Swinderen, Naotsugu Tsuchiya, Kavan Modi

Abstract: We apply techniques from the field of computational mechanics to evaluate the statistical complexity of neural recording data from fruit flies. First, we connect statistical complexity to the flies' level of conscious arousal, which is manipulated by general anesthesia (isoflurane). We show that the complexity of even single channel time series data decreases under anesthesia. The observed differe… ▽ More We apply techniques from the field of computational mechanics to evaluate the statistical complexity of neural recording data from fruit flies. First, we connect statistical complexity to the flies' level of conscious arousal, which is manipulated by general anesthesia (isoflurane). We show that the complexity of even single channel time series data decreases under anesthesia. The observed difference in complexity between the two states of conscious arousal increases as higher orders of temporal correlations are taken into account. We then go on to show that, in addition to reducing complexity, anesthesia also modulates the informational structure between the forward- and reverse-time neural signals. Specifically, using three distinct notions of temporal asymmetry we show that anesthesia reduces temporal asymmetry on information-theoretic and information-geometric grounds. In contrast to prior work, our results show that: (1) Complexity differences can emerge at very short timescales and across broad regions of the fly brain, thus heralding the macroscopic state of anesthesia in a previously unforeseen manner, and (2) that general anesthesia also modulates the temporal asymmetry of neural signals. Together, our results demonstrate that anesthetized brains become both less structured and more reversible. △ Less

Submitted 2 June, 2020; v1 submitted 30 May, 2019; originally announced May 2019.

Comments: 14 pages, 6 figures. Comments welcome; Added time-reversal analysis, updated discussion, new figures (Fig. 5 & Fig. 6) and Tables (Tab. 1)

Journal ref: Phys. Rev. Research 2, 023219 (2020)

Time-reversed and coherently-enhanced memory: A single-mode quantum atom-optic memory without a cavity

Authors: Jesse L. Everett, Pierre Vernaz-Gris, Geoff T. Campbell, Aaron D. Tranter, Karun V. Paul, Anthony C. Leung, ** Koy Lam, Ben C. Buchler

Abstract: The efficiency of an ensemble-based optical quantum memory depends critically on the strength of the atom-light coupling. An optical cavity is an effective method to enhance atom-light coupling strength, with the drawback that cavities can be difficult to integrate into a memory setup. In this work we show coherent enhancement of atom-light coupling via an interference effect. The light to be abso… ▽ More The efficiency of an ensemble-based optical quantum memory depends critically on the strength of the atom-light coupling. An optical cavity is an effective method to enhance atom-light coupling strength, with the drawback that cavities can be difficult to integrate into a memory setup. In this work we show coherent enhancement of atom-light coupling via an interference effect. The light to be absorbed into the atomic ensemble is split and used to drive the atoms from opposite ends of the ensemble. We compare this method theoretically to a cavity enhanced scheme and present experimental results for our coherent enhancement in cold rubidium-87 atoms that show an efficiency of $72\pm5\%$ and a storage lifetime of $110\pm 10$ us. △ Less

Submitted 21 January, 2019; originally announced January 2019.

Journal ref: Phys. Rev. A 98, 063846 (2018)

High-performance Raman memory with spatio-temporal reversal

Authors: Pierre Vernaz-Gris, Aaron D. Tranter, Jesse L. Everett, Anthony C. Leung, Karun V. Paul, Geoff T. Campbell, ** Koy Lam, Ben C. Buchler

Abstract: A number of techniques exist to use an ensemble of atoms as a quantum memory for light. Many of these propose to use backward retrieval as a way to improve the storage and recall efficiency. We report on a demonstration of an off-resonant Raman memory that uses backward retrieval to achieve an efficiency of $65\pm6\%$ at a storage time of one pulse duration. The memory has a characteristic decay t… ▽ More A number of techniques exist to use an ensemble of atoms as a quantum memory for light. Many of these propose to use backward retrieval as a way to improve the storage and recall efficiency. We report on a demonstration of an off-resonant Raman memory that uses backward retrieval to achieve an efficiency of $65\pm6\%$ at a storage time of one pulse duration. The memory has a characteristic decay time of 60 $μ$s, corresponding to a delay-bandwidth product of $160$. △ Less

Submitted 2 May, 2018; originally announced May 2018.

Journal ref: Optics Express Vol. 26, Issue 10, pp. 12424-12431 (2018)

Multiparameter optimisation of a magneto-optical trap using deep learning

Authors: Aaron D. Tranter, Harry J. Slatyer, Michael R. Hush, Anthony C. Leung, Jesse L. Everett, Karun V. Paul, Pierre Vernaz-Gris, ** Koy Lam, Ben C. Buchler, Geoff T. Campbell

Abstract: Many important physical processes have dynamics that are too complex to completely model analytically. Optimisation of such processes often relies on intuition, trial-and-error, or the construction of empirical models. Machine learning based on artificial neural networks has emerged as an efficient means to develop empirical models of complex systems. We implement a deep artificial neural network… ▽ More Many important physical processes have dynamics that are too complex to completely model analytically. Optimisation of such processes often relies on intuition, trial-and-error, or the construction of empirical models. Machine learning based on artificial neural networks has emerged as an efficient means to develop empirical models of complex systems. We implement a deep artificial neural network to optimise the magneto-optic cooling and trap** of neutral atomic ensembles. Cold atomic ensembles have become commonplace in laboratories around the world, however, many-body interactions give rise to complex dynamics that preclude precise analytic optimisation of the cooling and trap** process. The solution identified by machine learning is radically different to the smoothly varying adiabatic solutions currently used. Despite this, the solutions vastly outperform best known solutions producing higher optical densities. This may provide a pathway to a new understanding of the dynamics of the cooling and trap** processes in cold atomic ensembles. △ Less

Submitted 2 May, 2018; originally announced May 2018.

Journal ref: Nature Communications, 9, 4360 (2018)

Multivariate Location and Scatter Matrix Estimation Under Cellwise and Casewise Contamination

Authors: Andy Leung, Victor J. Yohai, Ruben H. Zamar

Abstract: We consider the problem of multivariate location and scatter matrix estimation when the data contain cellwise and casewise outliers. Agostinelli et al. (2015) propose a two-step approach to deal with this problem: first, apply a univariate filter to remove cellwise outliers and second, apply a generalized S-estimator to downweight casewise outliers. We improve this proposal in three main direction… ▽ More We consider the problem of multivariate location and scatter matrix estimation when the data contain cellwise and casewise outliers. Agostinelli et al. (2015) propose a two-step approach to deal with this problem: first, apply a univariate filter to remove cellwise outliers and second, apply a generalized S-estimator to downweight casewise outliers. We improve this proposal in three main directions. First, we introduce a consistent bivariate filter to be used in combination with the univariate filter in the first step. Second, we propose a new fast subsampling procedure to generate starting points for the generalized S-estimator in the second step. Third, we consider a non-monotonic weight function for the generalized S-estimator to better deal with casewise outliers in high dimension. A simulation study and real data example show that, unlike the original two-step procedure, the modified two-step approach performs and scales well for high dimension. Moreover, the modified procedure outperforms the original one and other state-of-the-art robust procedures under cellwise and casewise data contamination. △ Less

Submitted 25 December, 2016; v1 submitted 1 September, 2016; originally announced September 2016.

MSC Class: 62G35; 62G05; 62G20

Radial Star Formation Histories in Fifteen Nearby Galaxies

Authors: Daniel A. Dale, Gillian D. Beltz-Mohrmann, Arika A. Egan, Alan J. Hatlestad, Laura J. Herzog, Andrew S. Leung, Jacob N. McLane, Christopher Phenicie, Jareth S. Roberts, Kate L. Barnes, Mederic Boquien, Daniela Calzetti, David O. Cook, Henry A. Kobulnicky, Shawn M. Staudaher, Liese van Zee

Abstract: New deep optical and near-infrared imaging is combined with archival ultraviolet and infrared data for fifteen nearby galaxies mapped in the Spitzer Extended Disk Galaxy Exploration Science survey. These images are particularly deep and thus excellent for studying the low surface brightness outskirts of these disk-dominated galaxies with stellar masses ranging between 10^8 and 10^11 Msun. The spec… ▽ More New deep optical and near-infrared imaging is combined with archival ultraviolet and infrared data for fifteen nearby galaxies mapped in the Spitzer Extended Disk Galaxy Exploration Science survey. These images are particularly deep and thus excellent for studying the low surface brightness outskirts of these disk-dominated galaxies with stellar masses ranging between 10^8 and 10^11 Msun. The spectral energy distributions derived from this dataset are modeled to investigate the radial variations in the galaxy colors and star formation histories. Taken as a whole, the sample shows bluer and younger stars for larger radii until reversing near the optical radius, whereafter the trend is for redder and older stars for larger galacto-centric distances. These results are consistent with an inside-out disk formation scenario coupled with an old stellar outer disk population formed through radial migration and/or the cumulative history of minor mergers and accretions of satellite dwarf galaxies. However, these trends are quite modest and the variation from galaxy to galaxy is substantial. Additional data for a larger sample of galaxies are needed to confirm or dismiss these modest sample-wide trends. △ Less

Submitted 10 November, 2015; originally announced November 2015.

Comments: Accepted for publication in AJ

Bayesian Redshift Classification of Emission-line Galaxies with Photometric Equivalent Widths

Authors: Andrew S. Leung, Viviana Acquaviva, Eric Gawiser, Robin Ciardullo, Eiichiro Komatsu, A. I. Malz, Gregory R. Zeimann, Joanna S. Bridge, Niv Drory, John J. Feldmeier, Steven L. Finkelstein, Karl Gebhardt, Caryl Gronwall, Alex Hagen, Gary J. Hill, Donald P. Schneider

Abstract: We present a Bayesian approach to the redshift classification of emission-line galaxies when only a single emission line is detected spectroscopically. We consider the case of surveys for high-redshift Lyman-alpha-emitting galaxies (LAEs), which have traditionally been classified via an inferred rest-frame equivalent width (EW) greater than 20 angstrom. Our Bayesian method relies on known prior pr… ▽ More We present a Bayesian approach to the redshift classification of emission-line galaxies when only a single emission line is detected spectroscopically. We consider the case of surveys for high-redshift Lyman-alpha-emitting galaxies (LAEs), which have traditionally been classified via an inferred rest-frame equivalent width (EW) greater than 20 angstrom. Our Bayesian method relies on known prior probabilities in measured emission-line luminosity functions and equivalent width distributions for the galaxy populations, and returns the probability that an object in question is an LAE given the characteristics observed. This approach will be directly relevant for the Hobby-Eberly Telescope Dark Energy Experiment (HETDEX), which seeks to classify ~10^6 emission-line galaxies into LAEs and low-redshift [O II] emitters. For a simulated HETDEX catalog with realistic measurement noise, our Bayesian method recovers 86% of LAEs missed by the traditional EW > 20 angstrom cutoff over 2 < z < 3, outperforming the EW cut in both contamination and incompleteness. This is due to the method's ability to trade off between the two types of binary classification error by adjusting the stringency of the probability requirement for classifying an observed object as an LAE. In our simulations of HETDEX, this method reduces the uncertainty in cosmological distance measurements by 14% with respect to the EW cut, equivalent to recovering 29% more cosmological information. Rather than using binary object labels, this method enables the use of classification probabilities in large-scale structure analyses. It can be applied to narrowband emission-line surveys as well as upcoming large spectroscopic surveys including Euclid and WFIRST. △ Less

Submitted 21 April, 2016; v1 submitted 23 October, 2015; originally announced October 2015.

Comments: 16 pages, 7 figures, 5 tables, submitted to ApJ

Robust regression estimation and inference in the presence of cellwise and casewise contamination

Authors: Andy Leung, Hongyang Zhang, Ruben H. Zamar

Abstract: Cellwise outliers are likely to occur together with casewise outliers in modern data sets with relatively large dimension. Recent work has shown that traditional robust regression methods may fail for data sets in this paradigm. The proposed method, called three-step regression, proceeds as follows: first, it uses a consistent univariate filter to detect and eliminate extreme cellwise outliers; se… ▽ More Cellwise outliers are likely to occur together with casewise outliers in modern data sets with relatively large dimension. Recent work has shown that traditional robust regression methods may fail for data sets in this paradigm. The proposed method, called three-step regression, proceeds as follows: first, it uses a consistent univariate filter to detect and eliminate extreme cellwise outliers; second, it applies a robust estimator of multivariate location and scatter to the filtered data to down-weight casewise outliers; third, it computes robust regression coefficients from the estimates obtained in the second step. The three-step estimator is shown to be consistent and asymptotically normal at the central model under some assumptions on the tail distributions of the continuous covariates. The estimator is extended to handle both numerical and dummy covariates using an iterative algorithm. Extensive simulation results show that the three-step estimator is resilient to cellwise outliers. It also performs well under casewise contaminations when comparing with traditional high breakdown point estimators. △ Less

Submitted 25 December, 2016; v1 submitted 8 September, 2015; originally announced September 2015.

MSC Class: 62G35; 62G05; 62G20

A new measurement of antineutrino oscillation with the full detector configuration at Daya Bay

Authors: Daya Bay Collaboration, F. P. An, A. B. Balantekin, H. R. Band, M. Bishai, S. Blyth, I. Butorov, G. F. Cao, J. Cao, W. R. Cen, Y. L. Chan, J. F. Chang, L. C. Chang, Y. Chang, H. S. Chen, Q. Y. Chen, S. M. Chen, Y. X. Chen, Y. Chen, J. H. Cheng, J. Cheng, Y. P. Cheng, J. J. Cherwinka, M. C. Chu, J. P. Cummings , et al. (194 additional authors not shown)

Abstract: We report a new measurement of electron antineutrino disappearance using the fully-constructed Daya Bay Reactor Neutrino Experiment. The final two of eight antineutrino detectors were installed in the summer of 2012. Including the 404 days of data collected from October 2012 to November 2013 resulted in a total exposure of 6.9$\times$10$^5$ GW$_{\rm th}$-ton-days, a 3.6 times increase over our pre… ▽ More We report a new measurement of electron antineutrino disappearance using the fully-constructed Daya Bay Reactor Neutrino Experiment. The final two of eight antineutrino detectors were installed in the summer of 2012. Including the 404 days of data collected from October 2012 to November 2013 resulted in a total exposure of 6.9$\times$10$^5$ GW$_{\rm th}$-ton-days, a 3.6 times increase over our previous results. Improvements in energy calibration limited variations between detectors to 0.2%. Removal of six $^{241}$Am-$^{13}$C radioactive calibration sources reduced the background by a factor of two for the detectors in the experimental hall furthest from the reactors. Direct prediction of the antineutrino signal in the far detectors based on the measurements in the near detectors explicitly minimized the dependence of the measurement on models of reactor antineutrino emission. The uncertainties in our estimates of $\sin^{2}2θ_{13}$ and $|Δm^2_{ee}|$ were halved as a result of these improvements. Analysis of the relative antineutrino rates and energy spectra between detectors gave $\sin^{2}2θ_{13} = 0.084\pm0.005$ and $|Δm^{2}_{ee}|= (2.42\pm0.11) \times 10^{-3}$ eV$^2$ in the three-neutrino framework. △ Less

Submitted 10 September, 2015; v1 submitted 13 May, 2015; originally announced May 2015.

Comments: Updated to match final published version

Journal ref: Phys. Rev. Lett. 115, 111802 (2015)

Low/High Redshift Classification of Emission Line Galaxies in the HETDEX Survey

Authors: Viviana Acquaviva, Eric Gawiser, Andrew S. Leung, Mario R. Martin

Abstract: We discuss different methods to separate high- from low-redshift galaxies based on a combination of spectroscopic and photometric observations. Our baseline scenario is the Hobby-Eberly Telescope Dark Energy eXperiment (HETDEX) survey, which will observe several hundred thousand Lyman Alpha Emitting (LAE) galaxies at 1.9 < z < 3.5, and for which the main source of contamination is [OII]-emitting g… ▽ More We discuss different methods to separate high- from low-redshift galaxies based on a combination of spectroscopic and photometric observations. Our baseline scenario is the Hobby-Eberly Telescope Dark Energy eXperiment (HETDEX) survey, which will observe several hundred thousand Lyman Alpha Emitting (LAE) galaxies at 1.9 < z < 3.5, and for which the main source of contamination is [OII]-emitting galaxies at z < 0.5. Additional information useful for the separation comes from empirical knowledge of LAE and [OII] luminosity functions and equivalent width distributions as a function of redshift. We consider three separation techniques: a simple cut in equivalent width, a Bayesian separation method, and machine learning algorithms, including support vector machines. These methods can be easily applied to other surveys and used on simulated data in the framework of survey planning. △ Less

Submitted 10 November, 2014; originally announced November 2014.

Comments: 4 pages, accepted for publication in the Proceedings of the IAU Symposium 306: "Statistical Challenges in 21st Century Cosmology" (Lisbon, Portugal, May 2014)

PinView: Implicit Feedback in Content-Based Image Retrieval

Authors: Zakria Hussain, Arto Klami, Jussi Kujala, Alex P. Leung, Kitsuchart Pasupa, Peter Auer, Samuel Kaski, Jorma Laaksonen, John Shawe-Taylor

Abstract: This paper describes PinView, a content-based image retrieval system that exploits implicit relevance feedback collected during a search session. PinView contains several novel methods to infer the intent of the user. From relevance feedback, such as eye movements or pointer clicks, and visual features of images, PinView learns a similarity metric between images which depends on the current intere… ▽ More This paper describes PinView, a content-based image retrieval system that exploits implicit relevance feedback collected during a search session. PinView contains several novel methods to infer the intent of the user. From relevance feedback, such as eye movements or pointer clicks, and visual features of images, PinView learns a similarity metric between images which depends on the current interests of the user. It then retrieves images with a specialized online learning algorithm that balances the tradeoff between exploring new images and exploiting the already inferred interests of the user. We have integrated PinView to the content-based image retrieval system PicSOM, which enables applying PinView to real-world image databases. With the new algorithms PinView outperforms the original PicSOM, and in online experiments with real users the combination of implicit and explicit feedback gives the best results. △ Less

Submitted 2 October, 2014; originally announced October 2014.

Comments: 12 pages

Search for a Light Sterile Neutrino at Daya Bay

Authors: F. P. An, A. B. Balantekin, H. R. Band, W. Beriguete, M. Bishai, S. Blyth, I. Butorov, G. F. Cao, J. Cao, Y. L. Chan, J. F. Chang, L. C. Chang, Y. Chang, C. Chasman, H. Chen, Q. Y. Chen, S. M. Chen, X. Chen, X. Chen, Y. X. Chen, Y. Chen, Y. P. Cheng, J. J. Cherwinka, M. C. Chu, J. P. Cummings , et al. (210 additional authors not shown)

Abstract: A search for light sterile neutrino mixing was performed with the first 217 days of data from the Daya Bay Reactor Antineutrino Experiment. The experiment's unique configuration of multiple baselines from six 2.9~GW$_{\rm th}$ nuclear reactors to six antineutrino detectors deployed in two near (effective baselines 512~m and 561~m) and one far (1579~m) underground experimental halls makes it possib… ▽ More A search for light sterile neutrino mixing was performed with the first 217 days of data from the Daya Bay Reactor Antineutrino Experiment. The experiment's unique configuration of multiple baselines from six 2.9~GW$_{\rm th}$ nuclear reactors to six antineutrino detectors deployed in two near (effective baselines 512~m and 561~m) and one far (1579~m) underground experimental halls makes it possible to test for oscillations to a fourth (sterile) neutrino in the $10^{\rm -3}~{\rm eV}^{2} < |Δm_{41}^{2}| < 0.3~{\rm eV}^{2}$ range. The relative spectral distortion due to electron antineutrino disappearance was found to be consistent with that of the three-flavor oscillation model. The derived limits on $\sin^22θ_{14}$ cover the $10^{-3}~{\rm eV}^{2} \lesssim |Δm^{2}_{41}| \lesssim 0.1~{\rm eV}^{2}$ region, which was largely unexplored. △ Less

Submitted 8 October, 2014; v1 submitted 27 July, 2014; originally announced July 2014.

Comments: 7 pages, 4 figures

Journal ref: Phys. Rev. Lett. 113, 141802 (2014)

Independent Measurement of Theta13 via Neutron Capture on Hydrogen at Daya Bay

Authors: Daya Bay Collaboration, F. P. An, A. B. Balantekin, H. R. Band, W. Beriguete, M. Bishai, S. Blyth, I. Butorov, G. F. Cao, J. Cao, Y. L. Chan, J. F. Chang, L. C. Chang, Y. Chang, C. Chasman, H. Chen, Q. Y. Chen, S. M. Chen, X. Chen, X. Chen, Y. X. Chen, Y. Chen, Y. P. Cheng, J. J. Cherwinka, M. C. Chu , et al. (210 additional authors not shown)

Abstract: A new measurement of the $θ_{13}$ mixing angle has been obtained at the Daya Bay Reactor Neutrino Experiment via the detection of inverse beta decays tagged by neutron capture on hydrogen. The antineutrino events for hydrogen capture are distinct from those for gadolinium capture with largely different systematic uncertainties, allowing a determination independent of the gadolinium-capture result… ▽ More A new measurement of the $θ_{13}$ mixing angle has been obtained at the Daya Bay Reactor Neutrino Experiment via the detection of inverse beta decays tagged by neutron capture on hydrogen. The antineutrino events for hydrogen capture are distinct from those for gadolinium capture with largely different systematic uncertainties, allowing a determination independent of the gadolinium-capture result and an improvement on the precision of $θ_{13}$ measurement. With a 217-day antineutrino data set obtained with six antineutrino detectors and from six 2.9 GW$_{th}$ reactors, the rate deficit observed at the far hall is interpreted as $\sin^22θ_{13}=0.083\pm0.018$ in the three-flavor oscillation model. When combined with the gadolinium-capture result from Daya Bay, we obtain $\sin^22θ_{13}=0.089\pm0.008$ as the final result for the six-antineutrino-detector configuration of the Daya Bay experiment. △ Less

Submitted 23 July, 2014; v1 submitted 25 June, 2014; originally announced June 2014.

Journal ref: Phys. Rev. D 90, 071101 (2014)

Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination

Authors: Claudio Agostinelli, Andy Leung, Victor J. Yohai, Ruben H. Zamar

Abstract: Multivariate location and scatter matrix estimation is a cornerstone in multivariate data analysis. We consider this problem when the data may contain independent cellwise and casewise outliers. Flat data sets with a large number of variables and a relatively small number of cases are common place in modern statistical applications. In these cases global down-weighting of an entire case, as perfor… ▽ More Multivariate location and scatter matrix estimation is a cornerstone in multivariate data analysis. We consider this problem when the data may contain independent cellwise and casewise outliers. Flat data sets with a large number of variables and a relatively small number of cases are common place in modern statistical applications. In these cases global down-weighting of an entire case, as performed by traditional robust procedures, may lead to poor results. We highlight the need for a new generation of robust estimators that can efficiently deal with cellwise outliers and at the same time show good performance under casewise outliers. △ Less

Submitted 23 June, 2014; originally announced June 2014.

MSC Class: 62G35 (Primary); 62G05 (Secondary)

Spectral measurement of electron antineutrino oscillation amplitude and frequency at Daya Bay

Authors: Daya Bay Collaboration, F. P. An, A. B. Balantekin, H. R. Band, W. Beriguete, M. Bishai, S. Blyth, R. L. Brown, I. Butorov, G. F. Cao, J. Cao, R. Carr, Y. L. Chan, J. F. Chang, Y. Chang, C. Chasman, H. S. Chen, H. Y. Chen, S. J. Chen, S. M. Chen, X. C. Chen, X. H. Chen, Y. Chen, Y. X. Chen, Y. P. Cheng , et al. (214 additional authors not shown)

Abstract: A measurement of the energy dependence of antineutrino disappearance at the Daya Bay Reactor Neutrino Experiment is reported. Electron antineutrinos ($\overlineν_{e}$) from six $2.9$ GW$_{\rm th}$ reactors were detected with six detectors deployed in two near (effective baselines 512 m and 561 m) and one far (1579 m) underground experimental halls. Using 217 days of data, 41589 (203809 and 92912)… ▽ More A measurement of the energy dependence of antineutrino disappearance at the Daya Bay Reactor Neutrino Experiment is reported. Electron antineutrinos ($\overlineν_{e}$) from six $2.9$ GW$_{\rm th}$ reactors were detected with six detectors deployed in two near (effective baselines 512 m and 561 m) and one far (1579 m) underground experimental halls. Using 217 days of data, 41589 (203809 and 92912) antineutrino candidates were detected in the far hall (near halls). An improved measurement of the oscillation amplitude $\sin^{2}2θ_{13} = 0.090^{+0.008}_{-0.009} $ and the first direct measurement of the $\overlineν_{e}$ mass-squared difference $|Δm^{2}_{ee}|= (2.59_{-0.20}^{+0.19}) \times 10^{-3}\ {\rm eV}^2 $ is obtained using the observed $\overlineν_{e}$ rates and energy spectra in a three-neutrino framework. This value of $|Δm^{2}_{ee}|$ is consistent with $|Δm^{2}_{μμ}|$ measured by muon neutrino disappearance, supporting the three-flavor oscillation model. △ Less

Submitted 15 January, 2014; v1 submitted 24 October, 2013; originally announced October 2013.

Comments: As accepted for publication by Phys. Rev. Lett. including ancillary table

Journal ref: Phys. Rev. Lett. 112, 061801 (2014)

Dynamical studies of macroscopic superposition states: Phase engineering of controlled entangled number states of Bose-Einstein condensate in multiple wells

Authors: Mary Ann Leung, Khan W. Mahmud, William P. Reinhardt

Abstract: We provide a scheme for the generation of entangled number states of Bose-Einstein condensates in multiple wells with cyclic pairwise connectivity. The condensate ground state in a multiple well trap can self-evolve, when phase engineered with specific initial phase differences between the neighboring wells, to a macroscopic superposition state with controllable entanglement -- to multiple well ge… ▽ More We provide a scheme for the generation of entangled number states of Bose-Einstein condensates in multiple wells with cyclic pairwise connectivity. The condensate ground state in a multiple well trap can self-evolve, when phase engineered with specific initial phase differences between the neighboring wells, to a macroscopic superposition state with controllable entanglement -- to multiple well generalization of double well NOON states. We demonstrate through numerical simulations the creation of entangled states in three and four wells and then explore the creation of "larger" entangled states where there are either a larger number of particles in each well or a larger number of wells. The type of entanglement produced as the particle numbers, or interaction strength, increases changes in a novel and initially unexpected manner. △ Less

Submitted 13 June, 2011; v1 submitted 13 June, 2010; originally announced June 2010.

Comments: 13 pages, 14 figures

Journal ref: Molecular Physics 110, 801 (2012)

Traveling Wave Solutions for Lotka-Volterra System Re-Visited

Authors: Anthony W Leung, Xiaojie Hou, Wei Feng

Abstract: Using a new method of monotone iteration of a pair of smooth lower- and upper-solutions, the traveling wave solutions of the classical Lotka-Volterra system are shown to exist for a family of wave speeds. Such constructed upper and lower solution pair enables us to derive the explicit value of the minimal (critical) wave speed as well as the asymptotic rates of the wave solutions at infinities.… ▽ More Using a new method of monotone iteration of a pair of smooth lower- and upper-solutions, the traveling wave solutions of the classical Lotka-Volterra system are shown to exist for a family of wave speeds. Such constructed upper and lower solution pair enables us to derive the explicit value of the minimal (critical) wave speed as well as the asymptotic rates of the wave solutions at infinities. Furthermore, the traveling wave corresponding to each wave speed is unique modulo a translation of the origin. The stability of the traveling wave solutions with non-critical wave speed is also studied by spectral analysis of the linearized operator in exponentially weighted Banach spaces. △ Less

Submitted 9 September, 2009; originally announced September 2009.

Comments: 27 pages

MSC Class: 35B35

Quantum Entanglement Capacity with Classical Feedback

Authors: Alan W. Leung

Abstract: For any quantum discrete memoryless channel, we define a quantity called quantum entanglement capacity with classical feedback ($E_B$), and we show that this quantity lies between two other well-studied quantities. These two quantities - namely the quantum capacity assisted by two-way classical communication ($Q_2$) and the quantum capacity with classical feedback ($Q_B$) - are widely conjecture… ▽ More For any quantum discrete memoryless channel, we define a quantity called quantum entanglement capacity with classical feedback ($E_B$), and we show that this quantity lies between two other well-studied quantities. These two quantities - namely the quantum capacity assisted by two-way classical communication ($Q_2$) and the quantum capacity with classical feedback ($Q_B$) - are widely conjectured to be different: there exists quantum discrete memoryless channel for which $Q_2>Q_B$. We then present a general scheme to convert any quantum error-correcting codes into adaptive protocols for this newly-defined quantity of the quantum depolarizing channel, and illustrate with Cat (repetition) code and Shor code. We contrast the present notion with entanglement purification protocols by showing that whilst the Leung-Shor protocol can be applied directly, recurrence methods need to be supplemented with other techniques but at the same time offer a way to improve the aforementioned Cat code. For the quantum depolarizing channel, we prove a formula that gives lower bounds on the quantum capacity with classical feedback from any $E_B$ protocols. We then apply this formula to the $E_B$ protocols that we discuss to obtain new lower bounds on the quantum capacity with classical feedback of the quantum depolarizing channel. △ Less

Submitted 14 December, 2007; v1 submitted 19 September, 2007; originally announced September 2007.

Journal ref: Phys. Rev. A, vol.77, 012322 (2008)

Adaptive Entanglement Purification Protocols with Two-way Classical Communication

Authors: Alan W. Leung, Peter W. Shor

Abstract: We present a family of entanglement purification protocols that generalize four previous methods, namely the recurrence method, the modified recurrence method, and the two methods proposed by Maneva-Smolin and Leung-Shor. We will show that this family of protocols have improved yields over a wide range of initial fidelities F, and hence imply new lower bounds on the quantum capacity assisted by… ▽ More We present a family of entanglement purification protocols that generalize four previous methods, namely the recurrence method, the modified recurrence method, and the two methods proposed by Maneva-Smolin and Leung-Shor. We will show that this family of protocols have improved yields over a wide range of initial fidelities F, and hence imply new lower bounds on the quantum capacity assisted by two-way classical communication of the quantum depolarizing channel. In particular, the yields of these protocols are higher than the yield of universal hashing for F less than 0.993 and as F goes to 1. △ Less

Submitted 28 February, 2007; v1 submitted 15 February, 2007; originally announced February 2007.

Comments: 4 pages with 7 figures, references added

Entanglement purification with two-way classical communication

Authors: Alan W. Leung, Peter W. Shor

Abstract: We present an improved protocol for entanglement purification of bipartite mixed states. The protocol requires two-way classical communication and hence implies an improved lower bound on the quantum capacity with two-way classical communication of the quantum depolarizing channel. We present an improved protocol for entanglement purification of bipartite mixed states. The protocol requires two-way classical communication and hence implies an improved lower bound on the quantum capacity with two-way classical communication of the quantum depolarizing channel. △ Less

Submitted 21 February, 2007; v1 submitted 15 February, 2007; originally announced February 2007.

Comments: 10 pages with 4 figures, references added

Transversality for a Class of 3D Oscillators via Gyrostat Equations

Authors: J. L. Kuang, A. Y. T. Leung

Abstract: The class of 3D oscillators of interest includes the modified Brockett, Chua, Duffing, Ueda, modified Kapitaniak, generalized Lorenz, forced Lorenz, Rossler and YSVO oscillators. The homoclinic orbits of a symmetric gyrostat with wheels under torque-free motions are first exploited. The effects of the small external perturbation torques upon the rotational motions of the forced symmetric gyrosta… ▽ More The class of 3D oscillators of interest includes the modified Brockett, Chua, Duffing, Ueda, modified Kapitaniak, generalized Lorenz, forced Lorenz, Rossler and YSVO oscillators. The homoclinic orbits of a symmetric gyrostat with wheels under torque-free motions are first exploited. The effects of the small external perturbation torques upon the rotational motions of the forced symmetric gyrostat are investigated using the equation of the Melnikov integral. The real zeros of the Melnikov integral determine the transversality of the homoclinic orbits leading to a necessary condition for the existence of chaos. The equations of the 3D oscillators are then reduced to the Euler equations of the perturbed rotational motions of symmetric gyrostats. Algorithms are established to compute the required parameters for the gyrostat to represent the 3D oscillators at initiating the transversality. These parameters include the angular momenta of the wheels and the principal moments of inertia. The existence of real zeros of the Melnikov integral for the symmetric gyrostat implies the existence of transversal intersections of the perturbed solutions of the 3D oscillators. The 4th order Runge-Kutta algorithm is utilized to simulate and crosscheck the long-term chaotic behaviors of the dynamical systems. △ Less

Submitted 3 June, 2015; v1 submitted 5 March, 2004; originally announced March 2004.

Comments: This paper has been withdrawn by the author due to the main points of view of the article are obsolete

Phase engineering of squeezed states and controlled entangled number states of Bose-Einstein condensates in multiple wells

Authors: Khan W. Mahmud, Mary Ann Leung, William P. Reinhardt

Abstract: We provide a scheme for the generation of controlled entangled number states of Bose-Einstein condensates in multiple wells, and also provide a novel method for the creation of squeezed states without severe adiabatic constraints on barrier heights. The condensate in a multiple well trap can be evolved, starting with a specific initial phase difference between the neighboring wells, to a tunable… ▽ More We provide a scheme for the generation of controlled entangled number states of Bose-Einstein condensates in multiple wells, and also provide a novel method for the creation of squeezed states without severe adiabatic constraints on barrier heights. The condensate in a multiple well trap can be evolved, starting with a specific initial phase difference between the neighboring wells, to a tunable entangled state or a squeezed state. We propose a general formula for the initial phase difference between the neighboring wells that is valid for any number of wells, even and odd. △ Less

Submitted 15 March, 2005; v1 submitted 27 February, 2004; originally announced March 2004.

Comments: revised version, 5 pages, 4 figures

Dynamics of the Bose-Einstein condensate: quasi-one-dimension and beyond

Authors: Lincoln D. Carr, Mary Ann Leung, William P. Reinhardt

Abstract: It is shown that the quasi-one-dimensional Bose-Einstein condensate is experimentally accessible and rich in intriguing phenomena. We demonstrate numerically and analytically the existence, stability, and perturbation-induced dynamics of all types of stationary states of the quasi-one-dimensional nonlinear Schrodinger equation for both repulsive and attractive cases. Among our results are: the c… ▽ More It is shown that the quasi-one-dimensional Bose-Einstein condensate is experimentally accessible and rich in intriguing phenomena. We demonstrate numerically and analytically the existence, stability, and perturbation-induced dynamics of all types of stationary states of the quasi-one-dimensional nonlinear Schrodinger equation for both repulsive and attractive cases. Among our results are: the connection between stationary states and solitons; creation of vortices from such states; manipulation of such states with simple phase profiles; demonstration of the fragility of the condensate phase in response to shock; and a robust stabilization of the attractive Bose-Einstein condensate. △ Less

Submitted 17 April, 2000; originally announced April 2000.

Comments: 22 pages, 10 figures

Journal ref: J. Phys. B: At. Mol. Opt. Phys. 33, 3983 (2000)