-
Shallow Recurrent Decoder for Reduced Order Modeling of Plasma Dynamics
Authors:
J. Nathan Kutz,
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
Reduced order models are becoming increasingly important for rendering complex and multiscale spatio-temporal dynamics computationally tractable. The computational efficiency of such surrogate models is especially important for design, exhaustive exploration and physical understanding. Plasma simulations, in particular those applied to the study of ${\bf E}\times {\bf B}$ plasma discharges and tec…
▽ More
Reduced order models are becoming increasingly important for rendering complex and multiscale spatio-temporal dynamics computationally tractable. The computational efficiency of such surrogate models is especially important for design, exhaustive exploration and physical understanding. Plasma simulations, in particular those applied to the study of ${\bf E}\times {\bf B}$ plasma discharges and technologies, such as Hall thrusters, require substantial computational resources in order to resolve the multidimentional dynamics that span across wide spatial and temporal scales. Although high-fidelity computational tools are available to simulate such systems over limited conditions and in highly simplified geometries, simulations of full-size systems and/or extensive parametric studies over many geometric configurations and under different physical conditions are computationally intractable with conventional numerical tools. Thus, scientific studies and industrially oriented modeling of plasma systems, including the important ${\bf E}\times {\bf B}$ technologies, stand to significantly benefit from reduced order modeling algorithms. We develop a model reduction scheme based upon a {\em Shallow REcurrent Decoder} (SHRED) architecture. The scheme uses a neural network for encoding limited sensor measurements in time (sequence-to-sequence encoding) to full state-space reconstructions via a decoder network. Based upon the theory of separation of variables, the SHRED architecture is capable of (i) reconstructing full spatio-temporal fields with as little as three point sensors, even the fields that are not measured with sensor feeds but that are in dynamic coupling with the measured field, and (ii) forecasting the future state of the system using neural network roll-outs from the trained time encoding model.
△ Less
Submitted 20 May, 2024;
originally announced May 2024.
-
Spatially Optimized Compact Deep Metric Learning Model for Similarity Search
Authors:
Md. Farhadul Islam,
Md. Tanzim Reza,
Meem Arafat Manab,
Mohammad Rakibul Hasan Mahin,
Sarah Zabeen,
Jannatun Noor
Abstract:
Spatial optimization is often overlooked in many computer vision tasks. Filters should be able to recognize the features of an object regardless of where it is in the image. Similarity search is a crucial task where spatial features decide an important output. The capacity of convolution to capture visual patterns across various locations is limited. In contrast to convolution, the involution kern…
▽ More
Spatial optimization is often overlooked in many computer vision tasks. Filters should be able to recognize the features of an object regardless of where it is in the image. Similarity search is a crucial task where spatial features decide an important output. The capacity of convolution to capture visual patterns across various locations is limited. In contrast to convolution, the involution kernel is dynamically created at each pixel based on the pixel value and parameters that have been learned. This study demonstrates that utilizing a single layer of involution feature extractor alongside a compact convolution model significantly enhances the performance of similarity search. Additionally, we improve predictions by using the GELU activation function rather than the ReLU. The negligible amount of weight parameters in involution with a compact model with better performance makes the model very useful in real-world implementations. Our proposed model is below 1 megabyte in size. We have experimented with our proposed methodology and other models on CIFAR-10, FashionMNIST, and MNIST datasets. Our proposed method outperforms across all three datasets.
△ Less
Submitted 9 April, 2024;
originally announced April 2024.
-
Model, Analyze, and Comprehend User Interactions and Various Attributes within a Social Media Platform
Authors:
Md Kaykobad Reza,
S M Maksudul Alam,
Yiran Luo,
Youzhe Liu
Abstract:
How can we effectively model, analyze, and comprehend user interactions and various attributes within a social media platform based on post-comment relationship? In this study, we propose a novel graph-based approach to model and analyze user interactions within a social media platform based on post-comment relationship. We construct a user interaction graph from social media data and analyze it t…
▽ More
How can we effectively model, analyze, and comprehend user interactions and various attributes within a social media platform based on post-comment relationship? In this study, we propose a novel graph-based approach to model and analyze user interactions within a social media platform based on post-comment relationship. We construct a user interaction graph from social media data and analyze it to gain insights into community dynamics, user behavior, and content preferences. Our investigation reveals that while 56.05% of the active users are strongly connected within the community, only 0.8% of them significantly contribute to its dynamics. Moreover, we observe temporal variations in community activity, with certain periods experiencing heightened engagement. Additionally, our findings highlight a correlation between user activity and popularity showing that more active users are generally more popular. Alongside these, a preference for positive and informative content is also observed where 82.41% users preferred positive and informative content. Overall, our study provides a comprehensive framework for understanding and managing online communities, leveraging graph-based techniques to gain valuable insights into user behavior and community dynamics.
△ Less
Submitted 23 March, 2024;
originally announced March 2024.
-
Data-driven local operator finding for reduced-order modelling of plasma systems: II. Application to parametric dynamics
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll,
J. Nathan Kutz
Abstract:
Real-world systems often exhibit dynamics influenced by various parameters, either inherent or externally controllable, necessitating models capable of reliably capturing these parametric behaviors. Plasma technologies exemplify such systems. For example, phenomena governing global dynamics in Hall thrusters (a spacecraft propulsion technology) vary with various parameters, such as the "self-susta…
▽ More
Real-world systems often exhibit dynamics influenced by various parameters, either inherent or externally controllable, necessitating models capable of reliably capturing these parametric behaviors. Plasma technologies exemplify such systems. For example, phenomena governing global dynamics in Hall thrusters (a spacecraft propulsion technology) vary with various parameters, such as the "self-sustained electric field". In this Part II, following on the introduction of our novel data-driven local operator finding algorithm, Phi Method, in Part I, we showcase the method's effectiveness in learning parametric dynamics to predict system behavior across unseen parameter spaces. We present two adaptations: the "parametric Phi Method" and the "ensemble Phi Method", which are demonstrated through 2D fluid-flow-past-a-cylinder and 1D Hall-thruster-plasma-discharge problems. Comparative evaluation against parametric OPT-DMD in the fluid case demonstrates superior predictive performance of the parametric Phi Method. Across both test cases, parametric and ensemble Phi Method reliably recover governing parametric PDEs and offer accurate predictions over test parameters. Ensemble ROM analysis underscores Phi Method's robust learning of dominant dynamic coefficients with high confidence.
△ Less
Submitted 3 March, 2024;
originally announced March 2024.
-
Data-driven local operator finding for reduced-order modelling of plasma systems: I. Concept and verifications
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll,
J. Nathan Kutz
Abstract:
Reduced-order plasma models that can efficiently predict plasma behavior across various settings and configurations are highly sought after yet elusive. The demand for such models has surged in the past decade due to their potential to facilitate scientific research and expedite the development of plasma technologies. In line with the advancements in computational power and data-driven methods, we…
▽ More
Reduced-order plasma models that can efficiently predict plasma behavior across various settings and configurations are highly sought after yet elusive. The demand for such models has surged in the past decade due to their potential to facilitate scientific research and expedite the development of plasma technologies. In line with the advancements in computational power and data-driven methods, we introduce the "Phi Method" in this two-part article. Part I presents this novel algorithm, which employs constrained regression on a candidate term library informed by numerical discretization schemes to discover discretized systems of differential equations. We demonstrate Phi Method's efficacy in deriving reliable and robust reduced-order models (ROMs) for three test cases: the Lorenz attractor, flow past a cylinder, and a 1D Hall-thruster-representative plasma. Part II will delve into the method's application for parametric dynamics discovery. Our results show that ROMs derived from the Phi Method provide remarkably accurate predictions of systems' behavior, whether derived from steady-state or transient-state data. This underscores the method's potential for transforming plasma system modeling.
△ Less
Submitted 3 March, 2024;
originally announced March 2024.
-
A Computer Vision Based Approach for Stalking Detection Using a CNN-LSTM-MLP Hybrid Fusion Model
Authors:
Murad Hasan,
Shahriar Iqbal,
Md. Billal Hossain Faisal,
Md. Musnad Hossin Neloy,
Md. Tonmoy Kabir,
Md. Tanzim Reza,
Md. Golam Rabiul Alam,
Md Zia Uddin
Abstract:
Criminal and suspicious activity detection has become a popular research topic in recent years. The rapid growth of computer vision technologies has had a crucial impact on solving this issue. However, physical stalking detection is still a less explored area despite the evolution of modern technology. Nowadays, stalking in public places has become a common occurrence with women being the most aff…
▽ More
Criminal and suspicious activity detection has become a popular research topic in recent years. The rapid growth of computer vision technologies has had a crucial impact on solving this issue. However, physical stalking detection is still a less explored area despite the evolution of modern technology. Nowadays, stalking in public places has become a common occurrence with women being the most affected. Stalking is a visible action that usually occurs before any criminal activity begins as the stalker begins to follow, loiter, and stare at the victim before committing any criminal activity such as assault, kidnap**, rape, and so on. Therefore, it has become a necessity to detect stalking as all of these criminal activities can be stopped in the first place through stalking detection. In this research, we propose a novel deep learning-based hybrid fusion model to detect potential stalkers from a single video with a minimal number of frames. We extract multiple relevant features, such as facial landmarks, head pose estimation, and relative distance, as numerical values from video frames. This data is fed into a multilayer perceptron (MLP) to perform a classification task between a stalking and a non-stalking scenario. Simultaneously, the video frames are fed into a combination of convolutional and LSTM models to extract the spatio-temporal features. We use a fusion of these numerical and spatio-temporal features to build a classifier to detect stalking incidents. Additionally, we introduce a dataset consisting of stalking and non-stalking videos gathered from various feature films and television series, which is also used to train the model. The experimental results show the efficiency and dynamism of our proposed stalker detection system, achieving 89.58% testing accuracy with a significant improvement as compared to the state-of-the-art approaches.
△ Less
Submitted 5 February, 2024;
originally announced February 2024.
-
Cesàro summability of Taylor series in higher order weighted Dirichlet type spaces
Authors:
Soumitra Ghara,
Rajeev Gupta,
Md. Ramiz Reza
Abstract:
For a positive integer $m$ and a finite non-negative Borel measure $μ$ on the unit circle, we study the Hadamard multipliers of higher order weighted Dirichlet-type spaces $\mathcal H_{μ, m}$. We show that if $α>\frac{1}{2},$ then for any $f$ in $\mathcal H_{μ, m},$ the sequence of generalized Ces{à}ro sums $\{σ_n^α[f]\}$ converges to $f$. We further show that if $α=\frac{1}{2}$ then for the Dirac…
▽ More
For a positive integer $m$ and a finite non-negative Borel measure $μ$ on the unit circle, we study the Hadamard multipliers of higher order weighted Dirichlet-type spaces $\mathcal H_{μ, m}$. We show that if $α>\frac{1}{2},$ then for any $f$ in $\mathcal H_{μ, m},$ the sequence of generalized Ces{à}ro sums $\{σ_n^α[f]\}$ converges to $f$. We further show that if $α=\frac{1}{2}$ then for the Dirac delta measure supported at any point on the unit circle, the previous statement breaks down for every positive integer $m$.
△ Less
Submitted 31 December, 2023;
originally announced January 2024.
-
Labeling Indoor Scenes with Fusion of Out-of-the-Box Perception Models
Authors:
Yimeng Li,
Navid Rajabi,
Sulabh Shrestha,
Md Alimoor Reza,
Jana Kosecka
Abstract:
The image annotation stage is a critical and often the most time-consuming part required for training and evaluating object detection and semantic segmentation models. Deployment of the existing models in novel environments often requires detecting novel semantic classes not present in the training data. Furthermore, indoor scenes contain significant viewpoint variations, which need to be handled…
▽ More
The image annotation stage is a critical and often the most time-consuming part required for training and evaluating object detection and semantic segmentation models. Deployment of the existing models in novel environments often requires detecting novel semantic classes not present in the training data. Furthermore, indoor scenes contain significant viewpoint variations, which need to be handled properly by trained perception models. We propose to leverage the recent advancements in state-of-the-art models for bottom-up segmentation (SAM), object detection (Detic), and semantic segmentation (MaskFormer), all trained on large-scale datasets. We aim to develop a cost-effective labeling approach to obtain pseudo-labels for semantic segmentation and object instance detection in indoor environments, with the ultimate goal of facilitating the training of lightweight models for various downstream tasks. We also propose a multi-view labeling fusion stage, which considers the setting where multiple views of the scenes are available and can be used to identify and rectify single-view inconsistencies. We demonstrate the effectiveness of the proposed approach on the Active Vision dataset and the ADE20K dataset. We evaluate the quality of our labeling process by comparing it with human annotations. Also, we demonstrate the effectiveness of the obtained labels in downstream tasks such as object goal navigation and part discovery. In the context of object goal navigation, we depict enhanced performance using this fusion approach compared to a zero-shot baseline that utilizes large monolithic vision-language pre-trained models.
△ Less
Submitted 17 November, 2023;
originally announced November 2023.
-
GC-MVSNet: Multi-View, Multi-Scale, Geometrically-Consistent Multi-View Stereo
Authors:
Vibhas K. Vats,
Sripad Joshi,
David J. Crandall,
Md. Alimoor Reza,
Soon-heung Jung
Abstract:
Traditional multi-view stereo (MVS) methods rely heavily on photometric and geometric consistency constraints, but newer machine learning-based MVS methods check geometric consistency across multiple source views only as a post-processing step. In this paper, we present a novel approach that explicitly encourages geometric consistency of reference view depth maps across multiple source views at di…
▽ More
Traditional multi-view stereo (MVS) methods rely heavily on photometric and geometric consistency constraints, but newer machine learning-based MVS methods check geometric consistency across multiple source views only as a post-processing step. In this paper, we present a novel approach that explicitly encourages geometric consistency of reference view depth maps across multiple source views at different scales during learning (see Fig. 1). We find that adding this geometric consistency loss significantly accelerates learning by explicitly penalizing geometrically inconsistent pixels, reducing the training iteration requirements to nearly half that of other MVS methods. Our extensive experiments show that our approach achieves a new state-of-the-art on the DTU and BlendedMVS datasets, and competitive results on the Tanks and Temples benchmark. To the best of our knowledge, GC-MVSNet is the first attempt to enforce multi-view, multi-scale geometric consistency during learning.
△ Less
Submitted 21 December, 2023; v1 submitted 30 October, 2023;
originally announced October 2023.
-
Impact of Guidance and Interaction Strategies for LLM Use on Learner Performance and Perception
Authors:
Harsh Kumar,
Ilya Musabirov,
Mohi Reza,
Jiakai Shi,
Xinyuan Wang,
Joseph Jay Williams,
Anastasia Kuzminykh,
Michael Liut
Abstract:
Personalized chatbot-based teaching assistants can be crucial in addressing increasing classroom sizes, especially where direct teacher presence is limited. Large language models (LLMs) offer a promising avenue, with increasing research exploring their educational utility. However, the challenge lies not only in establishing the efficacy of LLMs but also in discerning the nuances of interaction be…
▽ More
Personalized chatbot-based teaching assistants can be crucial in addressing increasing classroom sizes, especially where direct teacher presence is limited. Large language models (LLMs) offer a promising avenue, with increasing research exploring their educational utility. However, the challenge lies not only in establishing the efficacy of LLMs but also in discerning the nuances of interaction between learners and these models, which impact learners' engagement and results. We conducted a formative study in an undergraduate computer science classroom (N=145) and a controlled experiment on Prolific (N=356) to explore the impact of four pedagogically informed guidance strategies on the learners' performance, confidence and trust in LLMs. Direct LLM answers marginally improved performance, while refining student solutions fostered trust. Structured guidance reduced random queries as well as instances of students copy-pasting assignment questions to the LLM. Our work highlights the role that teachers can play in sha** LLM-supported learning environments.
△ Less
Submitted 23 January, 2024; v1 submitted 12 October, 2023;
originally announced October 2023.
-
Robust Multimodal Learning with Missing Modalities via Parameter-Efficient Adaptation
Authors:
Md Kaykobad Reza,
Ashley Prater-Bennette,
M. Salman Asif
Abstract:
Multimodal learning seeks to utilize data from multiple sources to improve the overall performance of downstream tasks. It is desirable for redundancies in the data to make multimodal systems robust to missing or corrupted observations in some correlated modalities. However, we observe that the performance of several existing multimodal networks significantly deteriorates if one or multiple modali…
▽ More
Multimodal learning seeks to utilize data from multiple sources to improve the overall performance of downstream tasks. It is desirable for redundancies in the data to make multimodal systems robust to missing or corrupted observations in some correlated modalities. However, we observe that the performance of several existing multimodal networks significantly deteriorates if one or multiple modalities are absent at test time. To enable robustness to missing modalities, we propose a simple and parameter-efficient adaptation procedure for pretrained multimodal networks. In particular, we exploit modulation of intermediate features to compensate for the missing modalities. We demonstrate that such adaptation can partially bridge performance drop due to missing modalities and outperform independent, dedicated networks trained for the available modality combinations in some cases. The proposed adaptation requires extremely small number of parameters (e.g., fewer than 0.7% of the total parameters) and applicable to a wide range of modality combinations and tasks. We conduct a series of experiments to highlight the missing modality robustness of our proposed method on 5 different datasets for multimodal semantic segmentation, multimodal material segmentation, and multimodal sentiment analysis tasks. Our proposed method demonstrates versatility across various tasks and datasets, and outperforms existing methods for robust multimodal learning with missing modalities.
△ Less
Submitted 26 February, 2024; v1 submitted 5 October, 2023;
originally announced October 2023.
-
ABScribe: Rapid Exploration & Organization of Multiple Writing Variations in Human-AI Co-Writing Tasks using Large Language Models
Authors:
Mohi Reza,
Nathan Laundry,
Ilya Musabirov,
Peter Dushniku,
Zhi Yuan "Michael" Yu,
Kashish Mittal,
Tovi Grossman,
Michael Liut,
Anastasia Kuzminykh,
Joseph Jay Williams
Abstract:
Exploring alternative ideas by rewriting text is integral to the writing process. State-of-the-art Large Language Models (LLMs) can simplify writing variation generation. However, current interfaces pose challenges for simultaneous consideration of multiple variations: creating new variations without overwriting text can be difficult, and pasting them sequentially can clutter documents, increasing…
▽ More
Exploring alternative ideas by rewriting text is integral to the writing process. State-of-the-art Large Language Models (LLMs) can simplify writing variation generation. However, current interfaces pose challenges for simultaneous consideration of multiple variations: creating new variations without overwriting text can be difficult, and pasting them sequentially can clutter documents, increasing workload and disrupting writers' flow. To tackle this, we present ABScribe, an interface that supports rapid, yet visually structured, exploration and organization of writing variations in human-AI co-writing tasks. With ABScribe, users can swiftly modify variations using LLM prompts, which are auto-converted into reusable buttons. Variations are stored adjacently within text fields for rapid in-place comparisons using mouse-over interactions on a popup toolbar. Our user study with 12 writers shows that ABScribe significantly reduces task workload (d = 1.20, p < 0.001), enhances user perceptions of the revision process (d = 2.41, p < 0.001) compared to a popular baseline workflow, and provides insights into how writers explore variations using LLMs.
△ Less
Submitted 27 March, 2024; v1 submitted 29 September, 2023;
originally announced October 2023.
-
Effects of the applied fields' strength on the plasma behavior and processes in ExB plasma discharges of various propellants: II. Magnetic field
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
We present in this part II the effects of the magnetic field intensity on the properties of the plasma discharge and the underlying phenomena for different propellant's ion mass. The plasma setup represents a perpendicular configuration of the electric and magnetic fields, with the electric field along the axial direction and the magnetic field along the radial direction. The magnetic field intens…
▽ More
We present in this part II the effects of the magnetic field intensity on the properties of the plasma discharge and the underlying phenomena for different propellant's ion mass. The plasma setup represents a perpendicular configuration of the electric and magnetic fields, with the electric field along the axial direction and the magnetic field along the radial direction. The magnetic field intensity is changed from 5 to 30 mT, with 5 mT increments. The propellant gases are xenon, krypton, and argon. The simulations are carried out using a particle-in-cell (PIC) code based on the computationally efficient reduced-order PIC scheme. Similar to the observations in part I, we show that, across all propellants, the variation in the intensity of the magnetic field yields two distinct regimes of the plasma, where either the Modified Two Stream Instability (MTSI) or the Electron Cyclotron Drift Instability (ECDI) are present. Nonetheless, a third plasma regime is also observed for cases with moderate values of the magnetic field intensity (15 and 20 mT), in which the ECDI and the MTSI co-exist with comparable amplitudes. This described change in the plasma regime becomes clearly reflected in the radial distribution of the axial electron current density and the electron temperature anisotropy. Contrary to the effect of the electric field magnitude in part I, we observed here that the MTSI is absent at the relatively low magnetic field intensities (5 and 10 mT). At the relatively high magnitudes of the magnetic field (25 and 30 mT), the MTSI becomes strongly present, a long-wavelength wave mode develops, and the ECDI does not excite. An exception to this latter observation was noticed for xenon, for which the ECDI's presence persists up to the magnetic field peak value of 25 mT.
△ Less
Submitted 10 September, 2023;
originally announced September 2023.
-
Effects of the applied fields' strength on the plasma behavior and processes in ExB plasma discharges of various propellants: I. Electric field
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
We present, in this two-part article, an extensive study on the influence that the magnitudes of the applied electric (E) and magnetic (B) fields have on a collisionless plasma discharge of xenon, krypton, and argon in a 2D radial-azimuthal configuration with perpendicular orientation of the fields. The dependency of the behavior and the underlying processes of ExB discharges on the strength of el…
▽ More
We present, in this two-part article, an extensive study on the influence that the magnitudes of the applied electric (E) and magnetic (B) fields have on a collisionless plasma discharge of xenon, krypton, and argon in a 2D radial-azimuthal configuration with perpendicular orientation of the fields. The dependency of the behavior and the underlying processes of ExB discharges on the strength of electromagnetic field and ion mass has not yet been studied in depth and in a manner that can distinguish the role of each individual factor. This has been, on the one hand, due to the significant computational cost of conventional high-fidelity particle-in-cell (PIC) codes that do not allow for extensive simulations over a broad parameter space within practical timeframes. On the other hand, the experimental efforts have been limited, in part, by the measurements' spatial and temporal resolution. In this sense, the notably reduced computational cost of the reduced-order PIC scheme enables to numerically cast light on the parametric variations in various aspects of the physics of ExB discharges, such as high resolution spatial-temporal map**s of the plasma instabilities. In part I of the article, we focus on the effects of the E-field intensity. We demonstrate that the intensity of the field determines two distinct plasma regimes, which are characterized by different dominant instability campaigns. At relatively low E-field magnitudes, the Modified Two Stream Instability (MTSI) is dominant, whereas, at relatively high E-field magnitudes, the MTSI is mitigated, and the Electron Cyclotron Drift Instability (ECDI) becomes dominant. These two regimes are identified for all studied propellants. Consequent to the change in the plasma regime, the radial distribution of the axial electron current density and the electron temperature anisotropy vary.
△ Less
Submitted 10 September, 2023;
originally announced September 2023.
-
MMSFormer: Multimodal Transformer for Material and Semantic Segmentation
Authors:
Md Kaykobad Reza,
Ashley Prater-Bennette,
M. Salman Asif
Abstract:
Leveraging information across diverse modalities is known to enhance performance on multimodal segmentation tasks. However, effectively fusing information from different modalities remains challenging due to the unique characteristics of each modality. In this paper, we propose a novel fusion strategy that can effectively fuse information from different modality combinations. We also propose a new…
▽ More
Leveraging information across diverse modalities is known to enhance performance on multimodal segmentation tasks. However, effectively fusing information from different modalities remains challenging due to the unique characteristics of each modality. In this paper, we propose a novel fusion strategy that can effectively fuse information from different modality combinations. We also propose a new model named Multi-Modal Segmentation TransFormer (MMSFormer) that incorporates the proposed fusion strategy to perform multimodal material and semantic segmentation tasks. MMSFormer outperforms current state-of-the-art models on three different datasets. As we begin with only one input modality, performance improves progressively as additional modalities are incorporated, showcasing the effectiveness of the fusion block in combining useful information from diverse input modalities. Ablation studies show that different modules in the fusion block are crucial for overall model performance. Furthermore, our ablation studies also highlight the capacity of different input modalities to improve performance in the identification of different types of materials. The code and pretrained models will be made available at https://github.com/csiplab/MMSFormer.
△ Less
Submitted 7 April, 2024; v1 submitted 7 September, 2023;
originally announced September 2023.
-
Influence of the magnetic field's curvature on the radial-azimuthal dynamics of a Hall thruster plasma discharge with different propellants
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
The topology of the applied magnetic field is an important design aspect of Hall thrusters. For modern Hall thrusters, the field topology most often features curved lines with a concave (negative) curvature upstream of the field peak and a convex (positive) curvature downstream. Additionally, the advent of the magnetic shielding technique has resulted in the design of Hall thrusters with non-conve…
▽ More
The topology of the applied magnetic field is an important design aspect of Hall thrusters. For modern Hall thrusters, the field topology most often features curved lines with a concave (negative) curvature upstream of the field peak and a convex (positive) curvature downstream. Additionally, the advent of the magnetic shielding technique has resulted in the design of Hall thrusters with non-conventional magnetic fields that exhibit high degrees of concavity upstream of the field's peak. We carry out a rigorous and detailed study of the effects that the magnetic field's curvature has on the plasma properties and the underlying processes in a 2D configuration representative of a Hall thruster's radial-azimuthal cross-section. The analyses are performed for plasma discharges of three propellants: xenon, krypton, and argon. For each propellant, we have carried out high-fidelity reduced-order particle-in-cell (PIC) simulations with various degrees of positive and negative curvatures of the magnetic field. Corresponding 1D radial PIC simulations were also performed for xenon to compare the observations between 1D and 2D simulations. We observed that there are distinct differences in the plasma phenomena between the cases with positive and negative field curvatures. The instability spectra in the cases of positive curvature is mostly dominated by the Electron Cyclotron Drift Instability, whereas the Modified Two Stream Instability is dominant in the negative-curvature cases. The distribution of the plasma properties, particularly the electron and ion temperatures, and the contribution of various mechanisms to electrons' cross-field transport showed notable variations with the field's curvature, especially between the positive and the negative values. Finally, the magnetic field curvature was observed to majorly influence the ion beam divergence along the radial and azimuthal coordinates.
△ Less
Submitted 1 September, 2023;
originally announced September 2023.
-
Dynamic Mode Decomposition for data-driven analysis and reduced-order modelling of ExB plasmas: II. dynamics forecasting
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll,
J. Nathan Kutz
Abstract:
In part I of the article, we demonstrated that a variant of the Dynamic Mode Decomposition (DMD) algorithm based on variable projection optimization, called Optimized DMD (OPT-DMD), enables a robust identification of the dominant spatiotemporally coherent modes underlying the data across various test cases representing different physical parameters in an ExB simulation configuration. As the OPT-DM…
▽ More
In part I of the article, we demonstrated that a variant of the Dynamic Mode Decomposition (DMD) algorithm based on variable projection optimization, called Optimized DMD (OPT-DMD), enables a robust identification of the dominant spatiotemporally coherent modes underlying the data across various test cases representing different physical parameters in an ExB simulation configuration. As the OPT-DMD can be constrained to produce stable reduced-order models (ROMs) by construction, in this paper, we extend the application of the OPT-DMD and investigate the capabilities of the linear ROM from this algorithm toward forecasting in time of the plasma dynamics in configurations representative of the radial-azimuthal and axial-azimuthal cross-sections of a Hall thruster and over a range of simulation parameters in each test case. The predictive capacity of the OPT-DMD ROM is assessed primarily in terms of short-term dynamics forecast or, in other words, for large ratios of training-to-test data. However, the utility of the ROM for long-term dynamics forecasting is also presented for an example case in the radial-azimuthal configuration. The model's predictive performance is heterogeneous across various test cases. Nonetheless, a remarkable predictiveness is observed in the test cases that do not exhibit highly transient behaviors. Moreover, in all investigated cases, the error between the ground-truth and the reconstructed data from the OPT-DMD ROM remains bounded over time within both the training and the test window. As a result, despite its limitation in terms of generalized applicability to all plasma conditions, the OPT-DMD is proven as a reliable method to develop low computational cost and highly predictive data-driven reduced-order models in systems with a quasi-periodic global evolution of the plasma state.
△ Less
Submitted 25 August, 2023;
originally announced August 2023.
-
Dynamic Mode Decomposition for data-driven analysis and reduced-order modelling of ExB plasmas: I. Extraction of spatiotemporally coherent patterns
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll,
J. Nathan Kutz
Abstract:
In this two-part article, we evaluate the utility and the generalizability of the Dynamic Mode Decomposition (DMD) algorithm for data-driven analysis and reduced-order modelling of plasma dynamics in cross-field ExB configurations. The DMD algorithm is an interpretable data-driven method that finds a best-fit linear model describing the time evolution of spatiotemporally coherent structures (patte…
▽ More
In this two-part article, we evaluate the utility and the generalizability of the Dynamic Mode Decomposition (DMD) algorithm for data-driven analysis and reduced-order modelling of plasma dynamics in cross-field ExB configurations. The DMD algorithm is an interpretable data-driven method that finds a best-fit linear model describing the time evolution of spatiotemporally coherent structures (patterns) in data. We have applied the DMD to extensive high-fidelity datasets generated using a particle-in-cell (PIC) code based on a cost-efficient reduced-order PIC scheme. In this part, we first provide an overview of the concept of DMD and its underpinning Proper Orthogonal and Singular Value Decomposition methods. Two of the main DMD variants are next introduced. We then present and discuss the results of the DMD application in terms of the identification and extraction of the dominant spatiotemporal modes from high-fidelity data over a range of simulation conditions. We demonstrate that the DMD variant based on variable projection optimization (OPT-DMD) outperforms the basic DMD method in identification of the modes underlying the data, leading to notably more reliable reconstruction of the ground-truth. Furthermore, we show in multiple test cases that the discrete frequency spectrum of OPT-DMD-extracted modes is consistent with the temporal spectrum from the Fast Fourier Transform of the data. This observation implies that the OPT-DMD augments the conventional spectral analyses by being able to uniquely reveal the spatial structure of the dominant modes in the frequency spectra, thus, yielding more accessible, comprehensive information on the spatiotemporal characteristics of the plasma phenomena.
△ Less
Submitted 25 August, 2023;
originally announced August 2023.
-
CGBA: Curvature-aware Geometric Black-box Attack
Authors:
Md Farhamdur Reza,
Ali Rahmati,
Tianfu Wu,
Huaiyu Dai
Abstract:
Decision-based black-box attacks often necessitate a large number of queries to craft an adversarial example. Moreover, decision-based attacks based on querying boundary points in the estimated normal vector direction often suffer from inefficiency and convergence issues. In this paper, we propose a novel query-efficient curvature-aware geometric decision-based black-box attack (CGBA) that conduct…
▽ More
Decision-based black-box attacks often necessitate a large number of queries to craft an adversarial example. Moreover, decision-based attacks based on querying boundary points in the estimated normal vector direction often suffer from inefficiency and convergence issues. In this paper, we propose a novel query-efficient curvature-aware geometric decision-based black-box attack (CGBA) that conducts boundary search along a semicircular path on a restricted 2D plane to ensure finding a boundary point successfully irrespective of the boundary curvature. While the proposed CGBA attack can work effectively for an arbitrary decision boundary, it is particularly efficient in exploiting the low curvature to craft high-quality adversarial examples, which is widely seen and experimentally verified in commonly used classifiers under non-targeted attacks. In contrast, the decision boundaries often exhibit higher curvature under targeted attacks. Thus, we develop a new query-efficient variant, CGBA-H, that is adapted for the targeted attack. In addition, we further design an algorithm to obtain a better initial boundary point at the expense of some extra queries, which considerably enhances the performance of the targeted attack. Extensive experiments are conducted to evaluate the performance of our proposed methods against some well-known classifiers on the ImageNet and CIFAR10 datasets, demonstrating the superiority of CGBA and CGBA-H over state-of-the-art non-targeted and targeted attacks, respectively. The source code is available at https://github.com/Farhamdur/CGBA.
△ Less
Submitted 6 August, 2023;
originally announced August 2023.
-
Using disturbance function for vibration analysis of a beam with an open edge crack
Authors:
Mousa Rezaee,
Saeed Lotfan,
Vahid A. Maleki
Abstract:
In this article, the model presented by Shen and Pierre to investigate the transverse vibration behavior of a simply supported beam has been revised. This is done by applying more realistic assumptions. The crack is modeled as a continuous disturbance and the disturbance function is provided based on fracture mechanics. Next, the natural frequencies corresponding to the model are extracted using t…
▽ More
In this article, the model presented by Shen and Pierre to investigate the transverse vibration behavior of a simply supported beam has been revised. This is done by applying more realistic assumptions. The crack is modeled as a continuous disturbance and the disturbance function is provided based on fracture mechanics. Next, the natural frequencies corresponding to the model are extracted using the Galerkin method. The effect of crack parameters on the vibration behavior of the cracked beam is investigated. The obtained results show that the natural frequencies of the beam decrease with increasing crack depth. At the end, the obtained results are compared with the experimental results. The results show that the presented model is improved compared to previous models and predicts the vibration behavior of cracked beams with better accuracy for different crack parameters.
△ Less
Submitted 12 March, 2023;
originally announced May 2023.
-
Ab-initio investigation of the physical properties of BaAgAs Dirac semimetal and its possible thermo-mechanical and optoelectronic applications
Authors:
A. S. M. Muhasin Reza,
S. H. Naqib
Abstract:
BaAgAs is a ternary Dirac semimetal which can be tuned across a number of topological orders. In this study we have investigated the bulk physical properties of BaAgAs using density functional theory based computations. Most of the results presented in this work are novel. The optimized structural parameters are in good agreement with previous results. The elastic constants indicate that BaAgAs is…
▽ More
BaAgAs is a ternary Dirac semimetal which can be tuned across a number of topological orders. In this study we have investigated the bulk physical properties of BaAgAs using density functional theory based computations. Most of the results presented in this work are novel. The optimized structural parameters are in good agreement with previous results. The elastic constants indicate that BaAgAs is mechanically stable and brittle in nature. The compound is moderately hard and possesses fair degree of machinability. There is significant mechanical/elastic anisotropy in BaAgAs. The Debye temperature of the compound is medium and the phonon thermal conductivity and melting temperature are moderate as well. The bonding character is mixed with notable covalent contribution. The electronic band structure calculations reveal clear semimetallic behavior with a Dirac node at the Fermi level. BaAgAs has a small ellipsoidal Fermi surface centered at the G-point of the Brillouin zone. The phonon dispersion curves show dynamical stability. There is a clear phonon band gap between the acoustic and the optical branches. The energy dependent optical constants conform to the band structure calculations. The compound is an efficient absorber of the ultraviolet light and has potential to be used as an anti-reflection coating. Optical anisotropy of BaAgAs is moderate. The computed repulsive Coulomb pseudopotential is low indicating that the electronic correlations in this compound are not strong.
△ Less
Submitted 12 May, 2023;
originally announced May 2023.
-
Reduced-order particle-in-cell simulations of a high-power magnetically shielded Hall thruster
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll,
Antonio Piragino,
Tommaso Andreussi,
Tommaso Misuri
Abstract:
High-power magnetically shielded Hall thrusters have recently emerged to meet the needs of the next-generation space missions. Even though a few such thrusters are currently undergoing their late-stage development campaigns, many unanswered questions yet exist concerning the behavior and evolution of the plasma in these large-size thrusters that feature an unconventional magnetic field topology. N…
▽ More
High-power magnetically shielded Hall thrusters have recently emerged to meet the needs of the next-generation space missions. Even though a few such thrusters are currently undergoing their late-stage development campaigns, many unanswered questions yet exist concerning the behavior and evolution of the plasma in these large-size thrusters that feature an unconventional magnetic field topology. Noting the complex, multi-dimensional nature of plasma processes in Hall thrusters, high-fidelity particle-in-cell simulations are optimal tools to study the intricate plasma behavior. Nonetheless, the significant computational cost of traditional PIC schemes renders simulating high-power thrusters without any physics-altering speed-up factors unfeasible. Thus, in this article, we demonstrate the applicability of the novel reduced-order PIC scheme for a cost-efficient, self-consistent study of the high-power Hall thrusters by performing simulations of a 20 kW magnetically shielded Hall thruster along the axial-azimuthal and radial-azimuthal coordinates. The axial-azimuthal simulations are performed for three operating conditions in a rather simplified representation of the thruster's inherently 3D configuration. Nevertheless, we resolved self-consistently an unprecedented 650 us of the discharge evolution without any ad-hoc electron mobility model, capturing several breathing cycles and approximating the experimental performance parameters with an accuracy of 70 to 80 % across the operating conditions. The radial-azimuthal simulations casted further light on the evolution of the azimuthal instabilities and the resulting variations in the electrons' cross-field mobility and the plasma-wall interactions. Particularly, we observed the development of a long-wavelength, relatively low-frequency wave mode near the exit plane of the thruster's channel that induces a notable electron transport.
△ Less
Submitted 13 April, 2023;
originally announced April 2023.
-
Effects of the neutral dynamics model on the particle-in-cell simulations of a Hall thruster plasma discharge
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll
Abstract:
The dynamics of the neutral atoms in Hall thrusters affect several plasma processes, from the ionization to the electrons' mobility. In the context of Hall thruster's particle-in-cell (PIC) modeling, the neutrals are often treated kinetically, similar to the plasma species, and their interactions with themselves and the ions are resolved using the Direct-Simulation Monte-Carlo (DSMC) algorithm. Ho…
▽ More
The dynamics of the neutral atoms in Hall thrusters affect several plasma processes, from the ionization to the electrons' mobility. In the context of Hall thruster's particle-in-cell (PIC) modeling, the neutrals are often treated kinetically, similar to the plasma species, and their interactions with themselves and the ions are resolved using the Direct-Simulation Monte-Carlo (DSMC) algorithm. However, the DSMC approach is computationally resource demanding. Therefore, modeling the neutrals as a 1D fluid has been also pursued in simulations that do not involve the radial coordinate and, hence, do not resolve the neutrals' radial expansion. In this article, we present an extensive study on the sensitivity of the PIC simulations of Hall thruster discharge to the model used for the neutral dynamics. We carried out 1D axial PIC simulations with various fluid and kinetic models of the neutrals as well as self-consistent quasi-2D axial-azimuthal simulations with different neutrals' fluid descriptions. Our results show that the predictions of the simulations in either 1D or 2D configurations are highly sensitive to the neutrals' model, and that different treatments of the neutrals change the spatiotemporal evolution of the discharge. Moreover, we observed that considering the ion-neutral collisions causes a significant variation in the neutral temperature, thus requiring that the neutrals' energy equation to be included as well in their fluid system of equations. Finally, the self-consistent axial-azimuthal simulations highlighted that a neutrals' model based on the continuity conservation equation only is not an appropriate choice and leads to physically unexpected high-frequency global discharge oscillations.
△ Less
Submitted 2 April, 2023;
originally announced April 2023.
-
AnoMalNet: Outlier Detection based Malaria Cell Image Classification Method Leveraging Deep Autoencoder
Authors:
Aminul Huq,
Md Tanzim Reza,
Shahriar Hossain,
Shakib Mahmud Dipto
Abstract:
Class imbalance is a pervasive issue in the field of disease classification from medical images. It is necessary to balance out the class distribution while training a model for decent results. However, in the case of rare medical diseases, images from affected patients are much harder to come by compared to images from non-affected patients, resulting in unwanted class imbalance. Various processe…
▽ More
Class imbalance is a pervasive issue in the field of disease classification from medical images. It is necessary to balance out the class distribution while training a model for decent results. However, in the case of rare medical diseases, images from affected patients are much harder to come by compared to images from non-affected patients, resulting in unwanted class imbalance. Various processes of tackling class imbalance issues have been explored so far, each having its fair share of drawbacks. In this research, we propose an outlier detection based binary medical image classification technique which can handle even the most extreme case of class imbalance. We have utilized a dataset of malaria parasitized and uninfected cells. An autoencoder model titled AnoMalNet is trained with only the uninfected cell images at the beginning and then used to classify both the affected and non-affected cell images by thresholding a loss value. We have achieved an accuracy, precision, recall, and F1 score of 98.49%, 97.07%, 100%, and 98.52% respectively, performing better than large deep learning models and other published works. As our proposed approach can provide competitive results without needing the disease-positive samples during training, it should prove to be useful in binary disease classification on imbalanced datasets.
△ Less
Submitted 20 February, 2024; v1 submitted 10 March, 2023;
originally announced March 2023.
-
DFT based investigation of bulk mechanical, thermophysical and optoelectronic properties of PbTaSe2 topological semimetal
Authors:
A. S. M. Muhasin Reza,
S. H. Naqib
Abstract:
PbTaSe2 is a non-centrosymmetric topological semimetal. In this work we have explored the structural, elastic, mechanical, bonding, electronic, acoustic, thermal, and optical properties of PbTaSe2. The electronic bond structure calculations confirm semi-metallic character. Fermi surface topology shows both electron and hole sheets. The single crystal elastic constants reveal that PbTaSe2 is elasti…
▽ More
PbTaSe2 is a non-centrosymmetric topological semimetal. In this work we have explored the structural, elastic, mechanical, bonding, electronic, acoustic, thermal, and optical properties of PbTaSe2. The electronic bond structure calculations confirm semi-metallic character. Fermi surface topology shows both electron and hole sheets. The single crystal elastic constants reveal that PbTaSe2 is elastically stable. The compound is soft, brittle, and highly machinable at the same time. It also possesses very high level of dry lubricity. Various anisotropy indicators suggest that PbTaSe2 is elastically anisotropic with layered character. The phonon dynamics has been investigated. Phonon dispersion plot shows that the compound is dynamically stable with a clear frequency gap between the acoustic and optical branches. The Debye temperature, phonon thermal conductivity, and melting temperature of PbTaSe2 is low. The compound has medium Grüneisen parameter. The bonding character is mainly dominated by ionic bonding with some metallic contribution. The optical parameters have been studied in detail. The optical spectra reveal metallic features. The compound reflects visible light very efficiently (reflectance above 60%). It is also an efficient absorber of the ultraviolet light. The compound exhibits significant optical anisotropy with respect to the polarization directions of the incident electric field.
△ Less
Submitted 2 March, 2023;
originally announced March 2023.
-
Bi-isometries reducing the hyper-ranges of the coordinates
Authors:
Sameer Chavan,
Md. Ramiz Reza
Abstract:
Let $(S_1, S_2)$ be a bi-isometry, that is, a pair of commuting isometries $S_1$ and $S_2$ on a complex Hilbert space $\mathscr H.$ By the von Neumann-Wold decomposition, the hyper-range $\mathscr H_\infty(S_1):=\cap_{n=0}^\infty S^n_1\mathscr H$ of $S_1$ reduces $S_1$ to a unitary operator. Although $\mathscr H_\infty(S_1)$ is an invariant subspace for $S_2,$ in general, $\mathscr H_\infty(S_1)$…
▽ More
Let $(S_1, S_2)$ be a bi-isometry, that is, a pair of commuting isometries $S_1$ and $S_2$ on a complex Hilbert space $\mathscr H.$ By the von Neumann-Wold decomposition, the hyper-range $\mathscr H_\infty(S_1):=\cap_{n=0}^\infty S^n_1\mathscr H$ of $S_1$ reduces $S_1$ to a unitary operator. Although $\mathscr H_\infty(S_1)$ is an invariant subspace for $S_2,$ in general, $\mathscr H_\infty(S_1)$ is not a reducing subspace for $S_2.$ We show that $\mathscr H_\infty(S_1)$ reduces $S_2$ to an isometry if and only if the subspaces $S_2(\ker S^*_1)$ and $\mathscr H_\infty(S_1)$ of $\mathscr H$ are orthogonal. Further, we describe all bi-isometries $(S_1, S_2)$ satisfying the orthogonality condition mentioned above.
△ Less
Submitted 10 February, 2023;
originally announced February 2023.
-
Semantically-informed Hierarchical Event Modeling
Authors:
Shubhashis Roy Dipta,
Mehdi Rezaee,
Francis Ferraro
Abstract:
Prior work has shown that coupling sequential latent variable models with semantic ontological knowledge can improve the representational capabilities of event modeling approaches. In this work, we present a novel, doubly hierarchical, semi-supervised event modeling framework that provides structural hierarchy while also accounting for ontological hierarchy. Our approach consists of multiple layer…
▽ More
Prior work has shown that coupling sequential latent variable models with semantic ontological knowledge can improve the representational capabilities of event modeling approaches. In this work, we present a novel, doubly hierarchical, semi-supervised event modeling framework that provides structural hierarchy while also accounting for ontological hierarchy. Our approach consists of multiple layers of structured latent variables, where each successive layer compresses and abstracts the previous layers. We guide this compression through the injection of structured ontological knowledge that is defined at the type level of events: importantly, our model allows for partial injection of semantic knowledge and it does not depend on observing instances at any particular level of the semantic ontology. Across two different datasets and four different evaluation metrics, we demonstrate that our approach is able to out-perform the previous state-of-the-art approaches by up to 8.5%, demonstrating the benefits of structured and semantic hierarchical knowledge for event modeling.
△ Less
Submitted 30 May, 2023; v1 submitted 20 December, 2022;
originally announced December 2022.
-
Effects of magnetic field gradient and secondary electron emission on instabilities and transport in an ExB plasma configuration
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
Today, partially magnetized low-temperature plasmas (LTP) in an ExB configuration, where the applied magnetic field is perpendicular to the self-consistent electric field, have important industrial applications. Hall thrusters, a type of electrostatic plasma propulsion, are one of the main LTP technologies whose advancement is hindered by the not-fully-understood underlying physics of operation, p…
▽ More
Today, partially magnetized low-temperature plasmas (LTP) in an ExB configuration, where the applied magnetic field is perpendicular to the self-consistent electric field, have important industrial applications. Hall thrusters, a type of electrostatic plasma propulsion, are one of the main LTP technologies whose advancement is hindered by the not-fully-understood underlying physics of operation, particularly, with respect to the plasma instabilities and the associated electron cross-field transport. The development of Hall thrusters with unconventional magnetic field topologies has imposed further questions regarding the instabilities' characteristics and the electrons' dynamics in these modern cross-field configurations. Accordingly, we present in this effort a series of studies on the influence of four factors on the plasma processes in the radial-azimuthal coordinates of a Hall thruster, namely, the magnetic field gradient, Secondary Electron Emission, electron-neutral collisions, and plasma number density. The studies are carried out using the reduced-order particle-in-cell (PIC) code developed by the authors. The setup of the radial-azimuthal simulations largely follows a well-defined benchmark case from the literature in which the magnetic field is oriented along the radius and a constant axial electric field is applied perpendicular to the simulation plane. The salient finding from our investigations is that, in the studied cases corresponding to elevated plasma densities, an inverse energy cascade leads to the formation of a long-wavelength, high-frequency azimuthal mode. Moreover, in the presence of strong magnetic field gradients, this mode is fully developed and induces a significant electron cross-field transport as well as a notable heating of the ion population.
△ Less
Submitted 9 December, 2022;
originally announced December 2022.
-
Verification of the generalized reduced-order particle-in-cell scheme in a radial-azimuthal ExB plasma configuration
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll
Abstract:
In this article, we present an in-depth verification of the generalized electrostatic reduced-order particle-in-cell (PIC) scheme in a cross electric and magnetic field configuration representative of a radial-azimuthal section of a Hall thruster. The setup of the simulations follows a well-established benchmark case. The main purpose of this effort is to demonstrate that our novel PIC scheme can…
▽ More
In this article, we present an in-depth verification of the generalized electrostatic reduced-order particle-in-cell (PIC) scheme in a cross electric and magnetic field configuration representative of a radial-azimuthal section of a Hall thruster. The setup of the simulations follows a well-established benchmark case. The main purpose of this effort is to demonstrate that our novel PIC scheme can reliably resolve the complex two-dimensional dynamics and interactions of the plasma instabilities in the radial-azimuthal coordinates of a Hall thruster at a fraction of the computational cost compared to full-2D PIC codes. To this end, we first present the benchmarking of our newly developed full-2D PIC code. Next, we provide an overview of the reduced-order PIC scheme and the resulting "quasi-2D" code, specifying that the degree of order reduction in the quasi-2D PIC is defined in terms of the number of "regions" along the simulation's directions used to divide the computational domain. We compare the predictions of the quasi-2D simulation in various approximation degrees of the 2D problem against our full-2D simulation results. We show that, by increasing the number of regions in the Q2D simulations, the quasi-2D results converge to the 2D ones. Nonetheless, we also highlight that a quasi-2D simulation that provides a factor of 5 reduction in the computational cost resolves the underlying physical processes in an almost indistinguishable manner with respect to the full-2D simulation and incurs a maximum error of only about 10 % in the ion number density and about 5 % in the electron temperature.
△ Less
Submitted 29 November, 2022;
originally announced November 2022.
-
Explainable AI based Glaucoma Detection using Transfer Learning and LIME
Authors:
Touhidul Islam Chayan,
Anita Islam,
Eftykhar Rahman,
Md. Tanzim Reza,
Tasnim Sakib Apon,
MD. Golam Rabiul Alam
Abstract:
Glaucoma is the second driving reason for partial or complete blindness among all the visual deficiencies which mainly occurs because of excessive pressure in the eye due to anxiety or depression which damages the optic nerve and creates complications in vision. Traditional glaucoma screening is a time-consuming process that necessitates the medical professionals' constant attention, and even so t…
▽ More
Glaucoma is the second driving reason for partial or complete blindness among all the visual deficiencies which mainly occurs because of excessive pressure in the eye due to anxiety or depression which damages the optic nerve and creates complications in vision. Traditional glaucoma screening is a time-consuming process that necessitates the medical professionals' constant attention, and even so time to time due to the time constrains and pressure they fail to classify correctly that leads to wrong treatment. Numerous efforts have been made to automate the entire glaucoma classification procedure however, these existing models in general have a black box characteristics that prevents users from understanding the key reasons behind the prediction and thus medical practitioners generally can not rely on these system. In this article after comparing with various pre-trained models, we propose a transfer learning model that is able to classify Glaucoma with 94.71\% accuracy. In addition, we have utilized Local Interpretable Model-Agnostic Explanations(LIME) that introduces explainability in our system. This improvement enables medical professionals obtain important and comprehensive information that aid them in making judgments. It also lessen the opacity and fragility of the traditional deep learning models.
△ Less
Submitted 7 October, 2022;
originally announced October 2022.
-
Generalized reduced-order particle-in-cell scheme for Hall thruster modeling: concept and in-depth verification in the axial-azimuthal configuration
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
Reduced-order particle-in-cell (PIC) scheme is a novel modeling approach that enables computationally efficient electrostatic kinetic simulations of plasma. In our previous publications, we demonstrated the potentials of a preliminary implementation of this PIC scheme to resolve the multi-dimensional plasma processes in a Hall thruster in a manner close to traditional multi-dimensional PIC simulat…
▽ More
Reduced-order particle-in-cell (PIC) scheme is a novel modeling approach that enables computationally efficient electrostatic kinetic simulations of plasma. In our previous publications, we demonstrated the potentials of a preliminary implementation of this PIC scheme to resolve the multi-dimensional plasma processes in a Hall thruster in a manner close to traditional multi-dimensional PIC simulations. In this work, we first introduce a mathematically mature formulation for the dimensionality reduction of Poisson's equation, which enables the generalized "quasi-multi-dimensional" PIC scheme. We present the proof that the dimensionally reduced Poisson's equation converges to the multi-dimensional one in the limit. A reduced-dimension Poisson solver incorporating the dimensionality-reduction formulation is then verified for general 2D Poisson problems. Next, we present the results of several quasi-2D axial-azimuthal simulations performed using the generalized reduced-order electrostatic PIC scheme. We show that the improvements in the formulation of the reduced-order PIC notably augments its predictive capability, enabling the approach to reproduce quite accurately the results from reference full-2D PIC simulations in several conditions using low-order approximations of the 2D problem. Moreover, higher order approximations allow us to recover the same characteristics and behaviors reported in the literature regarding the azimuthal instabilities in Hall thrusters with simulations offering several factors reduction in computational cost. Finally, we discuss a series of sensitivity analyses, including the influence of the cathode boundary condition and the azimuthal domain length on the quasi-2D simulations' predictions. Accordingly, we outline in the conclusions the ongoing activities on the reduced-order scheme toward cost-effective three-dimensional studies of Hall thrusters' physics.
△ Less
Submitted 27 August, 2022;
originally announced August 2022.
-
Experimenting with Experimentation: Rethinking The Role of Experimentation in Educational Design
Authors:
Mohi Reza,
Akmar Chowdhury,
Aidan Li,
Mahathi Gandhamaneni,
Joseph Jay Williams
Abstract:
What if we take a broader view of what it means to run an education experiment? In this paper, we explore opportunities that arise when we think beyond the commonly-held notion that the purpose of an experiment is to either accept or reject a pre-defined hypothesis and instead, reconsider experimentation as a means to explore the complex design space of creating and improving instructional content…
▽ More
What if we take a broader view of what it means to run an education experiment? In this paper, we explore opportunities that arise when we think beyond the commonly-held notion that the purpose of an experiment is to either accept or reject a pre-defined hypothesis and instead, reconsider experimentation as a means to explore the complex design space of creating and improving instructional content. This is an approach we call experiment-inspired design. Then, to operationalize these ideas in a real-world experimentation venue, we investigate the implications of running a sequence of interventions teaching first-year students "meta-skills": transferable skills applicable to multiple areas of their lives, such as planning, and managing stress. Finally, using two examples as case studies for meta-skills interventions (stress-reappraisal and mental contrasting with implementation intentions), we reflect on our experiences with experiment-inspired design and share six preliminary lessons on how to use experimentation for design.
△ Less
Submitted 9 August, 2022;
originally announced August 2022.
-
Estimating Cosmological Constraints from Galaxy Cluster Abundance using Simulation-Based Inference
Authors:
Moonzarin Reza,
Yuanyuan Zhang,
Brian Nord,
Jason Poh,
Aleksandra Ciprijanovic,
Louis Strigari
Abstract:
Inferring the values and uncertainties of cosmological parameters in a cosmology model is of paramount importance for modern cosmic observations. In this paper, we use the simulation-based inference (SBI) approach to estimate cosmological constraints from a simplified galaxy cluster observation analysis. Using data generated from the Quijote simulation suite and analytical models, we train a machi…
▽ More
Inferring the values and uncertainties of cosmological parameters in a cosmology model is of paramount importance for modern cosmic observations. In this paper, we use the simulation-based inference (SBI) approach to estimate cosmological constraints from a simplified galaxy cluster observation analysis. Using data generated from the Quijote simulation suite and analytical models, we train a machine learning algorithm to learn the probability function between cosmological parameters and the possible galaxy cluster observables. The posterior distribution of the cosmological parameters at a given observation is then obtained by sampling the predictions from the trained algorithm. Our results show that the SBI method can successfully recover the truth values of the cosmological parameters within the 2σ limit for this simplified galaxy cluster analysis, and acquires similar posterior constraints obtained with a likelihood-based Markov Chain Monte Carlo method, the current state-of the-art method used in similar cosmological studies.
△ Less
Submitted 29 July, 2022;
originally announced August 2022.
-
A local Douglas formula for higher order weighted Dirichlet-type integrals
Authors:
Soumitra Ghara,
Rajeev Gupta,
Md. Ramiz Reza
Abstract:
We prove a local Douglas formula for higher order weighted Dirichlet-type integrals. With the help of this formula, we study the multiplier algebra of the associated higher order weighted Dirichlet-type spaces $\mathcal H_{\pmbμ},$ induced by an $m$-tuple $\pmb μ=(μ_1,\ldots,μ_{m})$ of finite non-negative Borel measures on the unit circle. In particular, it is shown that any weighted Dirichlet-typ…
▽ More
We prove a local Douglas formula for higher order weighted Dirichlet-type integrals. With the help of this formula, we study the multiplier algebra of the associated higher order weighted Dirichlet-type spaces $\mathcal H_{\pmbμ},$ induced by an $m$-tuple $\pmb μ=(μ_1,\ldots,μ_{m})$ of finite non-negative Borel measures on the unit circle. In particular, it is shown that any weighted Dirichlet-type space of order $m,$ for $m\geqslant 3,$ forms an algebra under pointwise product. We also prove that every non-zero closed $M_z$-invariant subspace of $\mathcal H_{\pmbμ},$ has codimension $1$ property if $m\geqslant 3$ or $μ_2$ is finitely supported. As another application of local Douglas formula obtained in this article, it is shown that for any $m\geqslant 2,$ weighted Dirichlet-type space of order $m$ does not coincide with any de Branges-Rovnyak space $\mathcal H(b)$ with equivalence of norms.
△ Less
Submitted 6 July, 2022;
originally announced July 2022.
-
Dimensionality reduction of Poisson's equation with application to particle-in-cell simulations of Hall thrusters
Authors:
Maryam Reza,
Farbod Faraji,
Aaron Knoll
Abstract:
In this article, we introduce a novel dimensionality reduction formulation for the Poisson's equation in the Vlasov-Poisson system that yields a reduced-order particle-in-cell scheme. This scheme allows a remarkable reduction in the computational cost of self-consistent kinetic simulations of Hall thrusters. The formulation of the dimensionality reduction approach, together with its verification f…
▽ More
In this article, we introduce a novel dimensionality reduction formulation for the Poisson's equation in the Vlasov-Poisson system that yields a reduced-order particle-in-cell scheme. This scheme allows a remarkable reduction in the computational cost of self-consistent kinetic simulations of Hall thrusters. The formulation of the dimensionality reduction approach, together with its verification for a general Poisson's problem, is presented. Moreover, we show the results of several "quasi-2D" axial-azimuthal simulations we performed with the conditions of a well-established 2D3V reference case. Comparison between the results of the quasi-2D simulations and the reference case revealed that our method is able to resolve accurately the axial distributions of the intensive plasma parameters. In addition, the characteristics of the azimuthal instabilities observed from the quasi-2D simulation are shown to be closely in line with those reported in the literature. Accordingly, the reduced-order PIC scheme is verified as a viable, low-computational-cost alternative to the traditional multi-dimensional particle-in-cell simulations of Hall thrusters.
△ Less
Submitted 28 June, 2022;
originally announced June 2022.
-
RevUp: Revise and Update Information Bottleneck for Event Representation
Authors:
Mehdi Rezaee,
Francis Ferraro
Abstract:
The existence of external (``side'') semantic knowledge has been shown to result in more expressive computational event models. To enable the use of side information that may be noisy or missing, we propose a semi-supervised information bottleneck-based discrete latent variable model. We reparameterize the model's discrete variables with auxiliary continuous latent variables and a light-weight hie…
▽ More
The existence of external (``side'') semantic knowledge has been shown to result in more expressive computational event models. To enable the use of side information that may be noisy or missing, we propose a semi-supervised information bottleneck-based discrete latent variable model. We reparameterize the model's discrete variables with auxiliary continuous latent variables and a light-weight hierarchical structure. Our model is learned to minimize the mutual information between the observed data and optional side knowledge that is not already captured by the new, auxiliary variables. We theoretically show that our approach generalizes past approaches, and perform an empirical case study of our approach on event modeling. We corroborate our theoretical results with strong empirical experiments, showing that the proposed method outperforms previous proposed approaches on multiple datasets.
△ Less
Submitted 14 February, 2023; v1 submitted 24 May, 2022;
originally announced May 2022.
-
Observation of Nonclassical Photon Statistics in Single-Bubble Sonoluminescence
Authors:
Mohammadreza Rezaee,
Yingwen Zhang,
James L. Harden,
Ebrahim Karimi
Abstract:
A cavitation bubble inside a liquid, under a specific set of conditions, can get trapped in an antinode of the ultrasonically driven standing wave and periodically emits visible photons (1,2). This conversion of sound to light phenomenon, known as sonoluminescence, can be seen with unaided eyes and occurs in multi- or single-bubble regimes. The sonoluminescence radiation spectrum analysis attribut…
▽ More
A cavitation bubble inside a liquid, under a specific set of conditions, can get trapped in an antinode of the ultrasonically driven standing wave and periodically emits visible photons (1,2). This conversion of sound to light phenomenon, known as sonoluminescence, can be seen with unaided eyes and occurs in multi- or single-bubble regimes. The sonoluminescence radiation spectrum analysis attributes a temperature of about 6,000 Kelvin to the bubble (3-5) -- close to the sun's surface temperature. The bubble dynamics, thermodynamics, and other physicochemical properties are well-explored and studied over the past decades (6-8). Notwithstanding that several theories such as the thermal (black-body radiation) model (9-11), shock wave theory (12), phase transition radiation (13,14), and vacuum quantum electrodynamics (15,16) -- were proposed, the underlying physics of sonoluminescence has hitherto remained a puzzle. Here, we experimentally investigated the photon number statistics of the emitted photons from single-bubble sonoluminescence (SBSL), employing two distinctly different techniques, i.e., measuring multiphoton correlations and photon number (using a photon-number resolving detector). Our findings show that the emitted photons from SBSL possess sub-Poissonian statistics, indicating the nonclassical nature of the SBSL. Our observation of sub-Poissonian statistics of emitted photons may help to explain the physics of SBSL. In addition, considering the importance of nonclassical light sources in quantum technologies, this discovery would be an exciting milestone in paving the route towards a bright quantum photonics source.
△ Less
Submitted 21 March, 2022;
originally announced March 2022.
-
Modulating Retroreflector Based Free Space Optical Link for UAV-to-Ground Communications
Authors:
Mohammad Taghi Dabiri,
Mohsen Rezaee,
Leila Mohammadi,
Farhang Javaherian,
Vahid Yazdanian,
Mazen Omar Hasna,
Murat Uysal
Abstract:
Weight reduction and low power consumption are key requirements in the next generation of unmanned aerial vehicle (UAV) networks. Employing modulating retro-reflector (MRR)-based free space optical (FSO) technology is an innovative technique for UAV-to-ground communication in order to reduce the payload weight and power consumption of UAVs which leads to increased maneuverability and flight time o…
▽ More
Weight reduction and low power consumption are key requirements in the next generation of unmanned aerial vehicle (UAV) networks. Employing modulating retro-reflector (MRR)-based free space optical (FSO) technology is an innovative technique for UAV-to-ground communication in order to reduce the payload weight and power consumption of UAVs which leads to increased maneuverability and flight time of UAV. In this paper, we consider an MRR-based FSO system for UAV-to-ground communication. We will show that the performance of the considered system is very sensitive to tracking errors. Therefore, to assess the benefits of MRR-based UAV deployment for FSO communications, the MRR-based UAV FSO channel is characterized by taking into account tracking system errors along with UAV's orientation fluctuations, link length, UAV's height, optical beam divergence angle, effective area of MRR, atmospheric turbulence and optical channel loss in the double-pass channels. To enable effective performance analysis, tractable and closed-form expressions are derived for probability density function of end-to-end signal to noise ratio, outage probability and bit error rate of the considered system under both weak-to-moderate and moderate-to-strong atmospheric turbulence conditions. The accuracy of the analytical expressions is verified by extensive simulations. Analytical results are then used to study the relationship between the optimal system design and tracking system errors.
△ Less
Submitted 26 February, 2022;
originally announced March 2022.
-
Enhancing one-dimensional particle-in-cell simulations to self-consistently resolve instability-induced electron transport in Hall thrusters
Authors:
Farbod Faraji,
Maryam Reza,
Aaron Knoll
Abstract:
The advent of high-power Hall thrusters and the increasing interest towards their use as a primary propulsion system for various missions have given a new boost to the efforts aiming at self-consistent predictive modeling of this thruster technology. In this article, we present a novel approach, which allows enhancing the predictive capability of one-dimensional Particle-in-Cell (PIC) simulations…
▽ More
The advent of high-power Hall thrusters and the increasing interest towards their use as a primary propulsion system for various missions have given a new boost to the efforts aiming at self-consistent predictive modeling of this thruster technology. In this article, we present a novel approach, which allows enhancing the predictive capability of one-dimensional Particle-in-Cell (PIC) simulations to self-consistently capture the wave-induced electron transport due to the azimuthal instabilities in Hall thrusters. The so-called "pseudo-2D" PIC scheme resulting from this approach is extensively tested in several operating conditions. The results are compared against a well-established 2D3V axial-azimuthal reference case in terms of the axial profiles of the time-averaged plasma properties, the azimuthal electric field fluctuations and their dispersion features, and the contributions of the force terms in the electron azimuthal momentum equation to the cross-field mobility. We have demonstrated that the pseudo-2D PIC provides a prediction of the above aspects that compares very closely in almost all conditions with those from the full-2D simulation. In addition, the sensitivity of the pseudo-2D simulation results to the numerical parameters associated with our approach is assessed in detail. The outcomes of these analyses have casted light on the next steps to further improve the approach.
△ Less
Submitted 16 May, 2022; v1 submitted 9 March, 2022;
originally announced March 2022.
-
A depleted and numerical model for pulsed Gaussian wave type II second harmonic generation
Authors:
Mostafa Mohammad Rezaee,
Mohammad Sabaeian,
Alireza Motazedian,
Fatemeh Sedaghat Jalil-Abadi
Abstract:
This paper presents a three-dimensional (3-D) and spatio-temporal dependent nonlinear wave model to explain the generation of pulsed Gaussian Second Harmonic Waves (SHW). We solve numerically three coupled equations describing the type II Second Harmonic Generation (SHG) in cylindrical KTP crystals under the assumption of fundamental waves depletion, two for ordinary and one for extraordinary Fund…
▽ More
This paper presents a three-dimensional (3-D) and spatio-temporal dependent nonlinear wave model to explain the generation of pulsed Gaussian Second Harmonic Waves (SHW). We solve numerically three coupled equations describing the type II Second Harmonic Generation (SHG) in cylindrical KTP crystals under the assumption of fundamental waves depletion, two for ordinary and one for extraordinary Fundamental Waves (FW). The results are attained by a homemade code written in FORTRAN. The results depict the efficiency of the SHG process with the conversion of FW energy to SHW energy, while SHW keeps the same Gaussian profile as FW. Furthermore, the results examine the effect of pulse energy, beam spot size, and therefore interaction length on SHW efficiency.
△ Less
Submitted 26 December, 2021;
originally announced December 2021.
-
Real Time Action Recognition from Video Footage
Authors:
Tasnim Sakib Apon,
Mushfiqul Islam Chowdhury,
MD Zubair Reza,
Arpita Datta,
Syeda Tan**a Hasan,
MD. Golam Rabiul Alam
Abstract:
Crime rate is increasing proportionally with the increasing rate of the population. The most prominent approach was to introduce Closed-Circuit Television (CCTV) camera-based surveillance to tackle the issue. Video surveillance cameras have added a new dimension to detect crime. Several research works on autonomous security camera surveillance are currently ongoing, where the fundamental goal is t…
▽ More
Crime rate is increasing proportionally with the increasing rate of the population. The most prominent approach was to introduce Closed-Circuit Television (CCTV) camera-based surveillance to tackle the issue. Video surveillance cameras have added a new dimension to detect crime. Several research works on autonomous security camera surveillance are currently ongoing, where the fundamental goal is to discover violent activity from video feeds. From the technical viewpoint, this is a challenging problem because analyzing a set of frames, i.e., videos in temporal dimension to detect violence might need careful machine learning model training to reduce false results. This research focuses on this problem by integrating state-of-the-art Deep Learning methods to ensure a robust pipeline for autonomous surveillance for detecting violent activities, e.g., kicking, punching, and slap**. Initially, we designed a dataset of this specific interest, which contains 600 videos (200 for each action). Later, we have utilized existing pre-trained model architectures to extract features, and later used deep learning network for classification. Also, We have classified our models' accuracy, and confusion matrix on different pre-trained architectures like VGG16, InceptionV3, ResNet50, Xception and MobileNet V2 among which VGG16 and MobileNet V2 performed better.
△ Less
Submitted 13 December, 2021;
originally announced December 2021.
-
Error Diagnosis of Deep Monocular Depth Estimation Models
Authors:
Jagpreet Chawla,
Nikhil Thakurdesai,
Anuj Godase,
Md Reza,
David Crandall,
Soon-Heung Jung
Abstract:
Estimating depth from a monocular image is an ill-posed problem: when the camera projects a 3D scene onto a 2D plane, depth information is inherently and permanently lost. Nevertheless, recent work has shown impressive results in estimating 3D structure from 2D images using deep learning. In this paper, we put on an introspective hat and analyze state-of-the-art monocular depth estimation models i…
▽ More
Estimating depth from a monocular image is an ill-posed problem: when the camera projects a 3D scene onto a 2D plane, depth information is inherently and permanently lost. Nevertheless, recent work has shown impressive results in estimating 3D structure from 2D images using deep learning. In this paper, we put on an introspective hat and analyze state-of-the-art monocular depth estimation models in indoor scenes to understand these models' limitations and error patterns. To address errors in depth estimation, we introduce a novel Depth Error Detection Network (DEDN) that spatially identifies erroneous depth predictions in the monocular depth estimation models. By experimenting with multiple state-of-the-art monocular indoor depth estimation models on multiple datasets, we show that our proposed depth error detection network can identify a significant number of errors in the predicted depth maps. Our module is flexible and can be readily plugged into any monocular depth prediction network to help diagnose its results. Additionally, we propose a simple yet effective Depth Error Correction Network (DECN) that iteratively corrects errors based on our initial error diagnosis.
△ Less
Submitted 15 November, 2021;
originally announced December 2021.
-
Quantum Face Recognition Protocol with Ghost Imaging
Authors:
Vahid Salari,
Dilip Paneru,
Erhan Saglamyurek,
Milad Ghadimi,
Moloud Abdar,
Mohammadreza Rezaee,
Mehdi Aslani,
Shabir Barzanjeh,
Ebrahim Karimi
Abstract:
Face recognition is one of the most ubiquitous examples of pattern recognition in machine learning, with numerous applications in security, access control, and law enforcement, among many others. Pattern recognition with classical algorithms requires significant computational resources, especially when dealing with high-resolution images in an extensive database. Quantum algorithms have been shown…
▽ More
Face recognition is one of the most ubiquitous examples of pattern recognition in machine learning, with numerous applications in security, access control, and law enforcement, among many others. Pattern recognition with classical algorithms requires significant computational resources, especially when dealing with high-resolution images in an extensive database. Quantum algorithms have been shown to improve the efficiency and speed of many computational tasks, and as such, they could also potentially improve the complexity of the face recognition process. Here, we propose a quantum machine learning algorithm for pattern recognition based on quantum principal component analysis (QPCA), and quantum independent component analysis (QICA). A novel quantum algorithm for finding dissimilarity in the faces based on the computation of trace and determinant of a matrix (image) is also proposed. The overall complexity of our pattern recognition algorithm is O(Nlog N) -- $N$ is the image dimension. As an input to these pattern recognition algorithms, we consider experimental images obtained from quantum imaging techniques with correlated photons, e.g. "interaction-free" imaging or "ghost" imaging. Interfacing these imaging techniques with our quantum pattern recognition processor provides input images that possess a better signal-to-noise ratio, lower exposures, and higher resolution, thus speeding up the machine learning process further. Our fully quantum pattern recognition system with quantum algorithm and quantum inputs promises a much-improved image acquisition and identification system with potential applications extending beyond face recognition, e.g., in medical imaging for diagnosing sensitive tissues or biology for protein identification.
△ Less
Submitted 19 October, 2021;
originally announced October 2021.
-
Theoretical and practical aspects of the design and production of synthetic holograms for transmission electron microscopy
Authors:
Paolo Rosi,
Federico Venturi,
Giacomo Medici,
Claudia Menozzi,
Gian Carlo Gazzadi,
Enzo Rotunno,
Stefano Frabboni,
Roberto Balboni,
Mohammadreza Rezaee,
Amir H. Tavabi,
Rafal E. Dunin-Borkowski,
Ebrahim Karimi,
Vincenzo Grillo
Abstract:
Beam sha** - the ability to engineer the phase and the amplitude of massive and massless particles - has long interested scientists working on communication, imaging and the foundations of quantum mechanics. In light optics, the sha** of electromagnetic waves (photons) can be achieved using techniques that include, but are not limited to, direct manipulation of the beam source (as in X-ray Fre…
▽ More
Beam sha** - the ability to engineer the phase and the amplitude of massive and massless particles - has long interested scientists working on communication, imaging and the foundations of quantum mechanics. In light optics, the sha** of electromagnetic waves (photons) can be achieved using techniques that include, but are not limited to, direct manipulation of the beam source (as in X-ray Free Electron Lasers (XFELs) and Synchrotrons), deformable mirrors, spatial light modulators, mode converters and holograms. The recent introduction of holographic masks for electrons provides new possibilities for electron beam sha**. Their fabrication has been made possible by advances in micrometric and nanometric device production using lithography and focused ion beam patterning. This article provides a tutorial on the generation, production and analysis of synthetic holograms for transmission electron microscopy. It begins with an introduction to synthetic holograms, outlining why they are useful for beam sha** to study material properties. It then focuses on the fabrication of the required devices from theoretical and experimental perspectives, with examples taken from both simulations and experimental results. Applications of synthetic electron holograms as aberration correctors, electron vortex generators and spatial mode sorters are then presented.
△ Less
Submitted 9 September, 2021;
originally announced September 2021.
-
Discriminative and Generative Transformer-based Models For Situation Entity Classification
Authors:
Mehdi Rezaee,
Kasra Darvish,
Gaoussou Youssouf Kebe,
Francis Ferraro
Abstract:
We re-examine the situation entity (SE) classification task with varying amounts of available training data. We exploit a Transformer-based variational autoencoder to encode sentences into a lower dimensional latent space, which is used to generate the text and learn a SE classifier. Test set and cross-genre evaluations show that when training data is plentiful, the proposed model can improve over…
▽ More
We re-examine the situation entity (SE) classification task with varying amounts of available training data. We exploit a Transformer-based variational autoencoder to encode sentences into a lower dimensional latent space, which is used to generate the text and learn a SE classifier. Test set and cross-genre evaluations show that when training data is plentiful, the proposed model can improve over the previous discriminative state-of-the-art models. Our approach performs disproportionately better with smaller amounts of training data, but when faced with extremely small sets (4 instances per label), generative RNN methods outperform transformers. Our work provides guidance for future efforts on SE and semantic prediction tasks, and low-label training regimes.
△ Less
Submitted 15 September, 2021;
originally announced September 2021.
-
The Cauchy dual subnormality problem via de Branges-Rovnyak spaces
Authors:
Sameer Chavan,
Soumitra Ghara,
Md Ramiz Reza
Abstract:
The Cauchy dual subnormality problem (for short, CDSP) asks whether the Cauchy dual of a $2$-isometry is subnormal. In this paper, we address this problem for cyclic $2$-isometries. In view of some recent developments in operator theory on function spaces (see \cite{AM, LGR}), one may recast CDSP as the problem of subnormality of the Cauchy dual $\mathscr M'_z$ of the multiplication operator…
▽ More
The Cauchy dual subnormality problem (for short, CDSP) asks whether the Cauchy dual of a $2$-isometry is subnormal. In this paper, we address this problem for cyclic $2$-isometries. In view of some recent developments in operator theory on function spaces (see \cite{AM, LGR}), one may recast CDSP as the problem of subnormality of the Cauchy dual $\mathscr M'_z$ of the multiplication operator $\mathscr M_z$ acting on a de Branges-Rovnyak space $\mathcal H(B),$ where $B$ is a vector-valued rational function. The main result of this paper characterizes the subnormality of $\mathscr M'_z$ on $\mathcal H(B)$ provided $B$ is a vector-valued rational function with simple poles. As an application, we provide affirmative solution to CDSP for the Dirichlet-type spaces $\mathscr D(μ)$ associated with measures $μ$ supported on two antipodal points of the unit circle.
△ Less
Submitted 19 April, 2021; v1 submitted 18 March, 2021;
originally announced March 2021.
-
Full-mode Characterisation of Correlated Photon Pairs Generated in Spontaneous Downconversion
Authors:
Alessio D'Errico,
Felix Hufnagel,
Filippo Miatto,
Mohammadreza Rezaee,
Ebrahim Karimi
Abstract:
Spontaneous parametric downconversion is the primary source to generate entangled photon pairs in quantum photonics laboratories. Depending on the experimental design, the generated photon pairs can be correlated in the frequency spectrum, polarisation, position-momentum, and spatial modes. Exploring the spatial modes' correlation has hitherto been limited to the polar coordinates' azimuthal angle…
▽ More
Spontaneous parametric downconversion is the primary source to generate entangled photon pairs in quantum photonics laboratories. Depending on the experimental design, the generated photon pairs can be correlated in the frequency spectrum, polarisation, position-momentum, and spatial modes. Exploring the spatial modes' correlation has hitherto been limited to the polar coordinates' azimuthal angle, and a few attempts to study Walsh modes radial states. Here, we study the full-mode correlation, on a Laguerre-Gauss basis, between photon pairs generated in a type-I crystal. Furthermore, we explore the effect of a structured pump beam possessing different spatial modes onto bi-photon spatial correlation. Finally, we use the capability to project over arbitrary spatial mode superpositions to perform the bi-photon state's full quantum tomography in a 16-dimensional subspace.
△ Less
Submitted 15 March, 2021;
originally announced March 2021.
-
Electronic Thermal Transport Measurement in Low-Dimensional Materials with Graphene Nonlocal Noise Thermometry
Authors:
Jonah Waissman,
Laurel E. Anderson,
Artem V. Talanov,
Zhongying Yan,
Young J. Shin,
Danial H. Najafabadi,
Mehdi Rezaee,
Xiaowen Feng,
Daniel G. Nocera,
Takashi Taniguchi,
Kenji Watanabe,
Brian Skinner,
Konstantin A. Matveev,
Philip Kim
Abstract:
In low-dimensional systems, the combination of reduced dimensionality, strong interactions, and topology has led to a growing number of many-body quantum phenomena. Thermal transport, which is sensitive to all energy-carrying degrees of freedom, provides a discriminating probe of emergent excitations in quantum materials and devices. However, thermal transport measurements in low dimensions are do…
▽ More
In low-dimensional systems, the combination of reduced dimensionality, strong interactions, and topology has led to a growing number of many-body quantum phenomena. Thermal transport, which is sensitive to all energy-carrying degrees of freedom, provides a discriminating probe of emergent excitations in quantum materials and devices. However, thermal transport measurements in low dimensions are dominated by the phonon contribution of the lattice, requiring an experimental approach to isolate the electronic thermal conductance. Here, we show how the measurement of nonlocal voltage fluctuations in a multiterminal device can reveal the electronic heat transported across a mesoscopic bridge made of low-dimensional materials. By using 2-dimensional graphene as a noise thermometer, we demonstrate quantitative electronic thermal conductance measurements of graphene and carbon nanotubes up to 70 K, achieving a precision of ~1% of the thermal conductance quantum at 5 K. Employing linear and nonlinear thermal transport, we observe signatures of long-range interaction-mediated energy transport in 1-dimensional electron systems, in agreement with a theoretical model. Our versatile nonlocal noise thermometry allows new experiments probing energy transport in emergent states of matter and devices in low dimensions.
△ Less
Submitted 3 January, 2022; v1 submitted 5 January, 2021;
originally announced January 2021.
-
DistB-Condo: Distributed Blockchain-based IoT-SDN Model for Smart Condominium
Authors:
Anichur Rahman,
Md. Jahidul Islam,
Ziaur Rahman,
Md. Mahfuz Reza,
Adnan Anwar,
M. A. Parvez Mahmud,
Mostofa Kamal Nasir,
Rafidah Md Noor
Abstract:
Condominium network refers to intra-organization networks, where smart buildings or apartments are connected and share resources over the network. Secured communication platform or channel has been highlighted as a key requirement for a reliable condominium which can be ensured by the utilization of the advanced techniques and platforms like Software-Defined Network (SDN), Network Function Virtual…
▽ More
Condominium network refers to intra-organization networks, where smart buildings or apartments are connected and share resources over the network. Secured communication platform or channel has been highlighted as a key requirement for a reliable condominium which can be ensured by the utilization of the advanced techniques and platforms like Software-Defined Network (SDN), Network Function Virtualization (NFV) and Blockchain (BC). These technologies provide a robust, and secured platform to meet all kinds of challenges, such as safety, confidentiality, flexibility, efficiency, and availability. This work suggests a distributed, scalable IoT-SDN with Blockchain-based NFV framework for a smart condominium (DistB-Condo) that can act as an efficient secured platform for a small community. Moreover, the Blockchain-based IoT-SDN with NFV framework provides the combined benefits of leading technologies. It also presents an optimized Cluster Head Selection (CHS) algorithm for selecting a Cluster Head (CH) among the clusters that efficiently saves energy. Besides, a decentralized and secured Blockchain approach has been introduced that allows more prominent security and privacy to the desired condominium network. Our proposed approach has also the ability to detect attacks in an IoT environment. Eventually, this article evaluates the performance of the proposed architecture using different parameters (e.g., throughput, packet arrival rate, and response time). The proposed approach outperforms the existing OF-Based SDN. DistB-Condo has better throughput on average, and the bandwidth (Mbps) much higher than the OF-Based SDN approach in the presence of attacks. Also, the proposed model has an average response time of 5% less than the core model.
△ Less
Submitted 17 December, 2020;
originally announced December 2020.
-
Study of Pluto's Atmosphere Based on 2020 Stellar Occultation Light Curve Results
Authors:
Atila Poro,
Farzaneh Ahangarani Farahani,
Majid Bahraminasr,
Maryam Hadizadeh,
Fatemeh Najafi kodini,
Maryam Rezaee,
Mahsa Seifi Gargari
Abstract:
On 6 Jun 2020, Pluto's stellar occultation was successfully observed at a ground-based observatory and Pluto's atmospheric parameters were investigated. We used an atmospheric model of Pluto (DO15), assuming a spherical and transparent pure N2 atmosphere. Using ray-tracing code the stellar occultation light curve was satisfactorily fitted to this model. We found that Pluto's atmospheric pressure a…
▽ More
On 6 Jun 2020, Pluto's stellar occultation was successfully observed at a ground-based observatory and Pluto's atmospheric parameters were investigated. We used an atmospheric model of Pluto (DO15), assuming a spherical and transparent pure N2 atmosphere. Using ray-tracing code the stellar occultation light curve was satisfactorily fitted to this model. We found that Pluto's atmospheric pressure at the reference radius of 1215 km is 6.72+_0.21 μbar. Our estimated pressure shows a continuation of the increasing pressure studied in 2016 consistent with a seasonal volatile transport model. We concluded that the N2 condensation processes in the Sputnik Planitia glacier are increasing due to the heating of the N2 ice in this basin. This study's result was shown on the diagram of the annual evolution of atmospheric pressure.
△ Less
Submitted 19 July, 2021; v1 submitted 9 November, 2020;
originally announced November 2020.