Search

The widespread implementation of urban surveillance systems has necessitated more sophisticated techniques for anomaly detection to ensure enhanced public safety. This paper presents a significant advancement in the field of anomaly detection through the application of Two-Stream Inflated 3D (I3D) Convolutional Networks. These networks substantially outperform traditional 3D Convolutional Networks (C3D) by more effectively extracting spatial and temporal features from surveillance videos, thus improving the precision of anomaly detection. Our research advances the field by implementing a weakly supervised learning framework based on Multiple Instance Learning (MIL), which uniquely conceptualizes surveillance videos as collections of 'bags' that contain instances (video clips). Each instance is innovatively processed through a ranking mechanism that prioritizes clips based on their potential to display anomalies. This novel strategy not only enhances the accuracy and precision of anomaly detection but also significantly diminishes the dependency on extensive manual annotations. Moreover, through meticulous optimization of model settings, including the choice of optimizer, our approach not only establishes new benchmarks in the performance of anomaly detection systems but also offers a scalable and efficient solution for real-world surveillance applications. This paper contributes significantly to the field of computer vision by delivering a more adaptable, efficient, and context-aware anomaly detection system, which is poised to redefine practices in urban surveillance., Comment: 11 pages, 8 figures

In this article, we look at the current bounds on the coupling strength of axion-like particles (ALPs) with two photons in the context of the Randall-Sundrum (RS) model. We relate the coupling strength to the compactification radius that governs the size of the extra dimension in the RS warped geometry model and show how the current bounds on the ALP can be used to derive appropriate constraints on the size of the extra fifth dimension in the RS model. We show that the resulting constraints fail to resolve the gauge hierarchy problem for light/ultralight ALPs and require a massive ALP of at least $m_{a} \gtrsim 0.1$ [GeV] to be relevant in the context of the hierarchy problem when the gauge field is in the bulk., Comment: 9 pages, 4 figures

This paper presents DEEGITS (Deep Learning Based Heterogeneous Traffic State Measurement), a comprehensive framework that leverages state-of-the-art convolutional neural network (CNN) techniques to accurately and rapidly detect vehicles and pedestrians, as well as to measure traffic states in challenging scenarios (i.e., congestion, occlusion). In this study, we enhance the training dataset through data fusion, enabling simultaneous detection of vehicles and pedestrians. Image preprocessing and augmentation are subsequently performed to improve the quality and quantity of the dataset. Transfer learning is applied on the YOLOv8 pretrained model to increase the model's capability to identify a diverse array of vehicles. Optimal hyperparameters are obtained using the Grid Search algorithm, with the Stochastic Gradient Descent (SGD) optimizer outperforming other optimizers under these settings. Extensive experimentation and evaluation demonstrate substantial accuracy within the detection framework, with the model achieving 0.794 mAP@0.5 on the validation set and 0.786 mAP@0.5 on the test set, surpassing previous benchmarks on similar datasets. The DeepSORT multi-object tracking algorithm is incorporated to track detected vehicles and pedestrians in this study. Finally, the framework is tested to measure heterogeneous traffic states in mixed traffic conditions. Two locations with differing traffic compositions and congestion levels are selected: one motorized-dominant location with moderate density and one non-motorized-dominant location with higher density. Errors are statistically insignificant for both cases, showing correlations from 0.99 to 0.88 and 0.91 to 0.97 for heterogeneous traffic flow and speed measurements, respectively., Comment: Submitted for presentation at the 103 rd Annual Meeting of Transportation Research Board and publication in Transportation Research Record: Journal of Transportation Research Board

In this work, we explore quantum energy teleportation (QET) protocols, focusing on their behavior at finite temperatures and in excited states. We analyze the role of entanglement as a resource for QET, particularly in thermal states, and compare the performance of QET across various initial states. We then introduce a method to extract ground-state energy through a protocol that employs only quantum measurements, local operations, and classical communication (LOCC), without requiring the ground state to be entangled. To illustrate this, we propose a minimal model comprising two interacting qubits. These findings indicate that, in addition to the established QET framework where entanglement serves as a resource, it is possible to extract energy from an unentangled ground state. This broadens the scope of QET's applicability across diverse quantum systems., Comment: 12 pages, 4 figures

We consider non-Hermitian tight-binding one-dimensional Hamiltonians and show that imposing a certain symmetry causes all eigenvalues to pair up and the corresponding eigenstates to coalesce in pairs. This Pairwise Coalescence (PC) is an enhanced version of an exceptional point -- the complete spectrum pairs up, not just one pair of eigenstates. The symmetry is that of reflection excluding the central two sites, and allowing flipping of non-reciprocal hoppings (``generalized off-center reflection symmetry''). Two simple examples of PC exist in the literature -- our construction encompasses these examples and extends them to a vast class of Hamiltonians. We study several families of such Hamiltonians, extend to cases of full-spectrum higher-order coalescences, and show how the PC point corresponds to amplified non-orthogonality of the eigenstates and enhanced loss of norm in time evolution., Comment: 15 pages

We establish global well-posedness for both the defocusing and focusing complex-valued modified Korteweg--de Vries equations on the real line in modulation spaces $M_p^{s,2}(\mathbb{R})$, for all $1\leq p<\infty$ and $0\leq s<3/2-1/p$. We will also show that such solutions admit global-in-time bounds in these spaces and that equicontinuous sets of initial data lead to equicontinuous ensembles of orbits. Indeed, such information forms a crucial part of our well-posedness argument., Comment: 23 pages

Incorporating advanced information and communication technologies into smart grids (SGs) offers substantial operational benefits while increasing vulnerability to cyber threats like false data injection (FDI) attacks. Current SG attack analysis tools predominantly employ formal methods or adversarial machine learning (ML) techniques with rule-based bad data detectors to analyze the attack space. However, these attack analytics either generate simplistic attack vectors detectable by the ML-based anomaly detection models (ADMs) or fail to identify critical attack vectors from complex controller dynamics in a feasible time. This paper introduces MISGUIDE, a novel defense-aware attack analytics designed to extract verifiable multi-time slot-based FDI attack vectors from complex SG load frequency control dynamics and ADMs, utilizing the Gurobi optimizer. MISGUIDE can identify optimal (maliciously triggering under/over frequency relays in minimal time) and stealthy attack vectors. Using real-world load data, we validate the MISGUIDE-identified attack vectors through real-time hardware-in-the-loop (OPALRT) simulations of the IEEE 39-bus system., Comment: 12 page journal

Large-scale cell microscopy screens are used in drug discovery and molecular biology research to study the effects of millions of chemical and genetic perturbations on cells. To use these images in downstream analysis, we need models that can map each image into a feature space that represents diverse biological phenotypes consistently, in the sense that perturbations with similar biological effects have similar representations. In this work, we present the largest foundation model for cell microscopy data to date, a new 1.9 billion-parameter ViT-G/8 MAE trained on over 8 billion microscopy image crops. Compared to a previous published ViT-L/8 MAE, our new model achieves a 60% improvement in linear separability of genetic perturbations and obtains the best overall performance on whole-genome biological relationship recall and replicate consistency benchmarks. Beyond scaling, we developed two key methods that improve performance: (1) training on a curated and diverse dataset; and, (2) using biologically motivated linear probing tasks to search across each transformer block for the best candidate representation of whole-genome screens. We find that many self-supervised vision transformers, pretrained on either natural or microscopy images, yield significantly more biologically meaningful representations of microscopy images in their intermediate blocks than in their typically used final blocks. More broadly, our approach and results provide insights toward a general strategy for successfully building foundation models for large-scale biological data., Comment: NeurIPS 2024 Foundation Models for Science Workshop (38th Conference on Neural Information Processing Systems). 18 pages, 7 figures

Large Language Models (LLMs) like GPT-3 and GPT-4 have emerged as groundbreaking innovations with capabilities that extend far beyond traditional AI applications. These sophisticated models, trained on massive datasets, can generate human-like text, respond to complex queries, and even write and interpret code. Their potential to revolutionize software development has captivated the software engineering (SE) community, sparking debates about their transformative impact. Through a critical analysis of technical strengths, limitations, real-world case studies, and future research directions, this paper argues that LLMs are not just reshaping how software is developed but are redefining the role of developers. While challenges persist, LLMs offer unprecedented opportunities for innovation and collaboration. Early adoption of LLMs in software engineering is crucial to stay competitive in this rapidly evolving landscape. This paper serves as a guide, helping developers, organizations, and researchers understand how to harness the power of LLMs to streamline workflows and acquire the necessary skills., Comment: 20 pages, 7 figures, 3 tables

Digital Enhanced Cordless Telecommunications 2020 New Radio (DECT-2020 NR) has garnered recognition as an alternative for cellular 5G technology in the internet of things industry. This paper presents a study centered around the analysis of the link distance performance in varying environments for DECT-2020 NR. The study extensively examines and analyzes received signal strength indicator and resulting path loss values in comparison with theoretical models, as well as packet success rates (SR) and signal-to-noise ratio against varying distances. The measurements show that with an SR of over 90%, an antenna height of 1.5 m, indoor link distances with a single device-to-device connection with 0 dBm transmission (TX) power can reach over 60 m in non-line-of-sight (NLOS) areas and up to 190 m in LOS areas with smaller -8 dBm TX power. Similarly, for outdoor use cases, link distances of over 600 m can be reached with +19 dBm TX power., Comment: Submitted for publication in EuCAP-2025 conference, 5 pages, 5 figures, and 4 tables

We present a novel experimental approach to detect dark matter by probing Yukawa interactions, commonly referred to as a fifth force, between dark matter and baryonic matter. Our method involves optically levitating nanoparticles within a Bessel-Gaussian beam to detect minute forces exerted by potential dark matter interaction with test masses. The non-diffracting properties of Bessel-Gaussian beams, combined with feedback cooling techniques, provide exceptional sensitivity to small perturbations in the motion of the nanoparticles. This setup allows for precise control over trapping conditions and enhances the detection sensitivity to forces on the order of \(10^{-18}\) N. We explore the parameter space of the Yukawa interaction, focusing on the coupling strength (\(\alpha\)) and interaction range (\(\lambda\)), and discuss the potential of this experiment to place new constraints on dark matter couplings, complementing existing direct detection methods., Comment: 22 pages, 7 figures, 1 table

Motivated by engineering applications such as resource allocation in networks and inventory systems, we consider average-reward Reinforcement Learning with unbounded state space and reward function. Recent works studied this problem in the actor-critic framework and established finite sample bounds assuming access to a critic with certain error guarantees. We complement their work by studying Temporal Difference (TD) learning with linear function approximation and establishing finite-time bounds with the optimal $\mathcal{O}\left(1/\epsilon^2\right)$ sample complexity. These results are obtained using the following general-purpose theorem for non-linear Stochastic Approximation (SA). Suppose that one constructs a Lyapunov function for a non-linear SA with certain drift condition. Then, our theorem establishes finite-time bounds when this SA is driven by unbounded Markovian noise under suitable conditions. It serves as a black box tool to generalize sample guarantees on SA from i.i.d. or martingale difference case to potentially unbounded Markovian noise. The generality and the mild assumption of the setup enables broad applicability of our theorem. We illustrate its power by studying two more systems: (i) We improve upon the finite-time bounds of $Q$-learning by tightening the error bounds and also allowing for a larger class of behavior policies. (ii) We establish the first ever finite-time bounds for distributed stochastic optimization of high-dimensional smooth strongly convex function using cyclic block coordinate descent.

Nonlinear optical materials are essential for the development of both nonlinear and quantum optics and have advanced recently from bulk crystals to integrated material platforms. In this Perspective, we provide an overview of the emerging InGaP $\chi^{(2)}$ nonlinear integrated photonics platform and its experimental achievements. With its exceptional $\chi^{(2)}$ nonlinearity and low optical losses, the epitaxial InGaP platform significantly enhances a wide range of second-order nonlinear optical effects, from second-harmonic generation to entangled photon pair sources, achieving efficiencies several orders of magnitude beyond the current state of the art. Moreover, the InGaP platform enables quantum nonlinear optics at the few- and single-photon levels via passive nonlinearities, which has broad implications for quantum information processing and quantum networking. We also examine the current limitations of the InGaP platform and propose potential solutions to fully unlock its capabilities.

We examine the impact of charged lepton Yukawa equilibration on leptogenesis during gravitational reheating. During the post-inflationary era, the inflaton field is assumed to oscillate around the minimum of a monomial potential, leading to the gravitational production of Standard Model (SM) particles, constituting the radiation bath. The heavy right-handed neutrinos (RHN), responsible for generating baryon asymmetry via leptogenesis, are also produced through graviton-mediated scattering of the homogeneous inflaton field and thermal bath, as well as from the the inverse decay of the bath particles. By considering both minimal and non-minimal gravitational contributions to SM, we demonstrate that flavour effects can be safely neglected in the minimal reheating scenario. However, with large non-minimal coupling, these effects may become important, depending on the choice of the RHN mass. We identify the corresponding viable parameter space that satisfies the observed baryon asymmetry in each case., Comment: 24 pages, 8 figures and 1 table

More than 3 kV breakdown voltage was demonstrated in monolithic bidirectional GaN HEMTs for the first time having potential applications in 1200V or 1700V-class novel power converters. The on resistance of the fabricated transistors was ~20 ohm.mm or ~11 mili ohm.cm^2. Breakdown voltage was optimized by utilizing two field plates in either side of the transistor and optimizing their geometry. Shorter first field plate lengths (less than 2 micron) resulted in higher breakdown voltage and the possible reason for this was discussed. The transistors had a steep subthreshold swing of 92 mV / dec. The on/off ratio was greater than 10^5 and it was limited by the tool capacity. The fabricated 3 kV transistor was benchmarked against the state-of-the-art monolithic bidirectional GaN HEMTs in the performance matrices of breakdown voltage and on resistance, that showed crucial progress., Comment: 4 pages, 7 figures

We propose a novel framework, Stable Diffusion-based Momentum Integrated Adversarial Examples (SD-MIAE), for generating adversarial examples that can effectively mislead neural network classifiers while maintaining visual imperceptibility and preserving the semantic similarity to the original class label. Our method leverages the text-to-image generation capabilities of the Stable Diffusion model by manipulating token embeddings corresponding to the specified class in its latent space. These token embeddings guide the generation of adversarial images that maintain high visual fidelity. The SD-MIAE framework consists of two phases: (1) an initial adversarial optimization phase that modifies token embeddings to produce misclassified yet natural-looking images and (2) a momentum-based optimization phase that refines the adversarial perturbations. By introducing momentum, our approach stabilizes the optimization of perturbations across iterations, enhancing both the misclassification rate and visual fidelity of the generated adversarial examples. Experimental results demonstrate that SD-MIAE achieves a high misclassification rate of 79%, improving by 35% over the state-of-the-art method while preserving the imperceptibility of adversarial perturbations and the semantic similarity to the original class label, making it a practical method for robust adversarial evaluation., Comment: 10 pages, 12 figures. To be published in IEEE TPS 2024 Proceedings. Code available on GitHub: https://github.com/nashrahhaque/SD-MIAE

Banks are important for the development of economies in any financial ecosystem through consumer and business loans. Lending, however, presents risks; thus, banks have to determine the applicant's financial position to reduce the probabilities of default. A number of banks have currently, therefore, adopted data analytics and state-of-the-art technology to arrive at better decisions in the process. The probability of payback is prescribed by a predictive modeling technique in which machine learning algorithms are applied. In this research project, we will apply several machine learning methods to further improve the accuracy and efficiency of loan approval processes. Our work focuses on the prediction of bank loan approval; we have worked on a dataset of 148,670 instances and 37 attributes using machine learning methods. The target property segregates the loan applications into "Approved" and "Denied" groups. various machine learning techniques have been used, namely, Decision Tree Categorization, AdaBoosting, Random Forest Classifier, SVM, and GaussianNB. Following that, the models were trained and evaluated. Among these, the best-performing algorithm was AdaBoosting, which achieved an incredible accuracy of 99.99%. The results therefore show how ensemble learning works effectively to improve the prediction skills of loan approval decisions. The presented work points to the possibility of achieving extremely accurate and efficient loan prediction models that provide useful insights for applying machine learning to financial domains., Comment: 10 pages, 18 figures, 6 tables

While cracking is a complex dynamics that involves material intrinsic properties like grain shape and size distribution, elastic properties of grain and cementing materials, and extrinsic properties of loading, in this work, the focus has been to check the dependence on the elastic properties of the bonding material. A 3-dimensional disordered system was constructed from spherical balls of varying radii that were chosen randomly from a log-normal distribution. The growth of micro-cracks with increasing compressive strain was monitored till the limit of the percolation crack. The two parameters varied were the bond stiffness constant and the bond strength of the material. Two distinct regimes of cracking rates were observed across a critical strain $\epsilon_{knee}$ that manifested as a knee in the cumulative crack-strain plot. The critical strain $\epsilon_{knee}$ and the strain at the percolation point $\epsilon_{perc}$ showed a power law dependence on the elastic property of the bond material. Individual micro-cracks were observed to grow sharply to a maximum value $N^{k_{b}}_{max}$, after which the number of new micro-cracks decreased, showing a long tail. The maximum $N^{k_{b}}_{max}$ was found to correspond to the strain $\epsilon_{knee}$, thus indicating that pre-$N^{k_{b}}_{max}$ cracking brittle, followed by ductile cracking behaviour of system. Lastly, we show that there exists a robust relation between $\epsilon_{knee}$ and $\epsilon_{perc}$ that is a power-law where the exponent is a function of the material elastic property. As $\epsilon_{knee}$ can be determined from acoustic signals associated with micro-cracks, our proposed relation can act as a warning towards critical strain resulting in crack percolation., Comment: 15 pages, 5 figures

We study the problem of computing the privacy parameters for DP machine learning when using privacy amplification via random batching and noise correlated across rounds via a correlation matrix $\textbf{C}$ (i.e., the matrix mechanism). Past work on this problem either only applied to banded $\textbf{C}$, or gave loose privacy parameters. In this work, we give a framework for computing near-exact privacy parameters for any lower-triangular, non-negative $\textbf{C}$. Our framework allows us to optimize the correlation matrix $\textbf{C}$ while accounting for amplification, whereas past work could not. Empirically, we show this lets us achieve smaller RMSE on prefix sums than the previous state-of-the-art (SOTA). We also show that we can improve on the SOTA performance on deep learning tasks. Our two main technical tools are (i) using Monte Carlo accounting to bypass composition, which was the main technical challenge for past work, and (ii) a "balls-in-bins" batching scheme that enables easy privacy analysis and is closer to practical random batching than Poisson sampling.

Integral geometry uses four geometric invariants -- the Minkowski functionals -- to characterize certain subsets of 3-dimensional space. The question was, how is the fluid flow in a 3-dimensional porous system related to these invariants? In this work, we systematically study the dependency of permeability on the geometrical characteristics of two categories of 3-dimensional porous systems generated: (i) stochastic and (ii) deterministic. For the stochastic systems, we investigated both normal and log-normal size distribution of grains. For the deterministic porous systems, we checked for a cubic and a hexagonal arrangement of grains of equal size. Our studies reveal that for any 3-dimensional porous system, ordered or disordered, permeability $k$ follows a unique scaling relation with the Minkowski functionals: (a) volume of the pore space, (b) integral mean curvature, (c) Euler Characteristic and (d) critical cross-sectional area of the pore space. The cubic and the hexagonal symmetrical systems formed the upper and lower bounds of the scaling relations, respectively. The disordered systems lay between these bounds. Moreover, we propose a combinatoric $F$ that weaves together the four Minkowski functionals and follows a power-law scaling with permeability. The scaling exponent is independent of particle size and distribution and has a universal value of $ 0.428$ for 3-dimensional porous systems built of spherical grains., Comment: 12 pages, 5 figures

Assume that $\,I\subseteq\mathbb{R}\,$ is an interval and $\,\beta:\,I\rightarrow\,I\,$ a strictly increasing and continuous function with a single fixed point $\,s_0\in I\,$, satisfying $\,(s_0-t)(\beta(t)-t)\leq 0\,$ for all $\,t\in I$, where the equality occurs only when $\,t=s_0$. Hamza et al. considered the general quantum operator, $\,D_{\beta}[f](t):=\displaystyle\frac{f\big(\beta(t)\big)-f(t)}{\beta(t)-t}\,$ when $\,t\neq s_0\,$ and $\,D_{\beta}[f](s_0):=f^{\prime}(s_0)\,$ when $\,t=s_0\,$. It generalizes the Jackson $\,q$-derivative operator $\,D_{q}\,$ as well as the Hahn (quantum derivative) operator, $\,D_{q,\omega}$. We obtained Gr\"uss type inequalities for its inverse operator, the $\beta$-integral. Furthermore, we introduced the concept of $\,\beta$-Riemann-Stieltjes integral and obtained Gr\"uss type inequalities associated with it.

This paper investigates the behavior of mutual information, entanglement negativity, and multipartite correlations in holographic RG flows, particularly during phase transitions. Mutual information provides a UV-finite measure of total correlations between subsystems, while entanglement negativity and multipartite correlations offer finer insights into quantum structures, especially near critical points. Through numerical simulations, we show that while mutual information remains relatively smooth, both entanglement negativity and multipartite correlations exhibit sharp changes near phase transitions. These results support the hypothesis that multipartite correlations play a dominant role in signaling critical phenomena in strongly coupled quantum systems., Comment: 21 pages, 5 figures

Numerous research studies have been conducted to enhance the resilience of cyber-physical systems (CPSs) by detecting potential cyber or physical disturbances. However, the development of scalable and optimal response measures under power system contingency based on fusing cyber-physical data is still in an early stage. To address this research gap, this paper introduces a power system response engine based on reinforcement learning (RL) and role and interaction discovery (RID) techniques. RL-RID-GridResponder is designed to automatically detect the contingency and assist with the decision-making process to ensure optimal power system operation. The RL-RID-GridResponder learns via an RL-based structure and achieves enhanced scalability by integrating an RID module with reduced action and state spaces. The applicability of RL-RID-GridResponder in providing scalable and optimal responses for CPSs is demonstrated on power systems in the context of Denial of Service (DoS) attacks. Moreover, simulations are conducted on a Volt-Var regulation problem using the augmented WSCC 9-bus and augmented IEEE 24-bus systems based on fused cyber and physical data sets. The results show that the proposed RL-RID-GridResponder can provide fast and accurate responses to ensure optimal power system operation under DoS and can extend to other system contingencies such as line outages and loss of loads.

This paper investigates the finite-temperature behavior of Conformal Field Theory (CFT) in Rindler vacuum, focusing on the relation between acceleration and thermality in quantum field theory. We illustrate how uniformly accelerated observers perceive the vacuum as a thermal state via Unruh effect, shedding light on the thermal properties of Rindler horizon. Through numerical simulations of the heat kernel, Unruh temperature, Planck distribution, and detector response, we demonstrate that acceleration enhances the thermal characteristics of quantum fields. These results provide important insights into horizon-induced thermality, with significant implications for black hole thermodynamics and quantum gravity., Comment: 15 pages, 4 figures

With the increasing adoption of AI-driven tools in software development, large language models (LLMs) have become essential for tasks like code generation, bug fixing, and optimization. Tools like ChatGPT, GitHub Copilot, and Codeium provide valuable assistance in solving programming challenges, yet their effectiveness remains underexplored. This paper presents a comparative study of ChatGPT, Codeium, and GitHub Copilot, evaluating their performance on LeetCode problems across varying difficulty levels and categories. Key metrics such as success rates, runtime efficiency, memory usage, and error-handling capabilities are assessed. GitHub Copilot showed superior performance on easier and medium tasks, while ChatGPT excelled in memory efficiency and debugging. Codeium, though promising, struggled with more complex problems. Despite their strengths, all tools faced challenges in handling harder problems. These insights provide a deeper understanding of each tool's capabilities and limitations, offering guidance for developers and researchers seeking to optimize AI integration in coding workflows.

The majority of medical images, especially those that resemble cells, have similar characteristics. These images, which occur in a variety of shapes, often show abnormalities in the organ or cell region. The convolution operation possesses a restricted capability to extract visual patterns across several spatial regions of an image. The involution process, which is the inverse operation of convolution, complements this inherent lack of spatial information extraction present in convolutions. In this study, we investigate how applying a single layer of involution prior to a convolutional neural network (CNN) architecture can significantly improve classification and segmentation performance, with a comparatively negligible amount of weight parameters. The study additionally shows how excessive use of involution layers might result in inaccurate predictions in a particular type of medical image. According to our findings from experiments, the strategy of adding only a single involution layer before a CNN-based model outperforms most of the previous works., Comment: 13 pages, 5 figures, 4 tables, preprint submitted to an Elsevier journal

This paper investigates the use of autoencoders and machine learning methods for detecting and analyzing quantum phase transitions in the Two-Component Bose-Hubbard Model. By leveraging deep learning models such as autoencoders, we investigate latent space representations, reconstruction error analysis, and cluster distance calculations to identify phase boundaries and critical points. The study is supplemented by dimensionality reduction techniques such as PCA and t-SNE for latent space visualization. The results demonstrate the potential of autoencoders to describe the dynamics of quantum phase transitions., Comment: 24 pages, 7 figures

Ensuring accessibility in mobile applications remains a significant challenge, particularly for visually impaired users who rely on screen readers. User interface icons are essential for navigation and interaction and often lack meaningful alt-text, creating barriers to effective use. Traditional deep learning approaches for generating alt-text require extensive datasets and struggle with the diversity and imbalance of icon types. More recent Vision Language Models (VLMs) require complete UI screens, which can be impractical during the iterative phases of app development. To address these issues, we introduce a novel method using Large Language Models (LLMs) to autonomously generate informative alt-text for mobile UI icons with partial UI data. By incorporating icon context, that include class, resource ID, bounds, OCR-detected text, and contextual information from parent and sibling nodes, we fine-tune an off-the-shelf LLM on a small dataset of approximately 1.4k icons, yielding IconDesc. In an empirical evaluation and a user study IconDesc demonstrates significant improvements in generating relevant alt-text. This ability makes IconDesc an invaluable tool for developers, aiding in the rapid iteration and enhancement of UI accessibility.

We propose a low energy model for simulating an analog black hole on an optical lattice using ultracold atoms. Assuming the validity of the holographic principle, we employ the Sachdev-Ye-Kitaev (SYK) model, which describes a system of randomly infinite range interacting fermions, also conjectured to be an exactly solvable UV-complete model for an extremal black hole in a higher dimensional Anti-de Sitter (AdS) dilaton gravity. At low energies, the SYK model exhibits an emergent conformal symmetry and is dual to the extremal black hole solution in near AdS2 spacetime. Furthermore, we show how the SYK maximally chaotic behaviour at large N limit, found to be dual to a gauge theory in higher dimensions, can also be employed as a non-trivial investigation tool for the holographic principle. The proposed setup is a theoretical platform to realize the SYK model with relevant exotic effects and behaviour at low energies as a highly non-trivial example of the AdS/CFT duality and a framework for studying black holes., Comment: 15 pages, 3 figures

Traffic routing is vital for the proper functioning of the Internet. As users and network traffic increase, researchers try to develop adaptive and intelligent routing algorithms that can fulfill various QoS requirements. Reinforcement Learning (RL) based routing algorithms have shown better performance than traditional approaches. We developed a QoS-aware, reusable RL routing algorithm, RLSR-Routing over SDN. During the learning process, our algorithm ensures loop-free path exploration. While finding the path for one traffic demand (a source destination pair with certain amount of traffic), RLSR-Routing learns the overall network QoS status, which can be used to speed up algorithm convergence when finding the path for other traffic demands. By adapting Segment Routing, our algorithm can achieve flow-based, source packet routing, and reduce communications required between SDN controller and network plane. Our algorithm shows better performance in terms of load balancing than the traditional approaches. It also has faster convergence than the non-reusable RL approach when finding paths for multiple traffic demands., Comment: 19 pages, 11 figures, Submitted to Elsevier Journal of Computer Network

Automatic Traffic Sign Recognition is paramount in modern transportation systems, motivating several research endeavors to focus on performance improvement by utilizing large-scale datasets. As the appearance of traffic signs varies across countries, curating large-scale datasets is often impractical; and requires efficient models that can produce satisfactory performance using limited data. In this connection, we present 'FUSED-Net', built-upon Faster RCNN for traffic sign detection, enhanced by Unfrozen Parameters, Pseudo-Support Sets, Embedding Normalization, and Domain Adaptation while reducing data requirement. Unlike traditional approaches, we keep all parameters unfrozen during training, enabling FUSED-Net to learn from limited samples. The generation of a Pseudo-Support Set through data augmentation further enhances performance by compensating for the scarcity of target domain data. Additionally, Embedding Normalization is incorporated to reduce intra-class variance, standardizing feature representation. Domain Adaptation, achieved by pre-training on a diverse traffic sign dataset distinct from the target domain, improves model generalization. Evaluating FUSED-Net on the BDTSD dataset, we achieved 2.4x, 2.2x, 1.5x, and 1.3x improvements of mAP in 1-shot, 3-shot, 5-shot, and 10-shot scenarios, respectively compared to the state-of-the-art Few-Shot Object Detection (FSOD) models. Additionally, we outperform state-of-the-art works on the cross-domain FSOD benchmark under several scenarios., Comment: 17 pages, 6 figures, 3 tables, submitted to IEEE Access for review

Version control systems (VCS) are essential for software development, yet centralized VCS present risks such as data loss, security breaches, and ownership disputes. While blockchain-based approaches to decentralized source code repository hosting have been explored, many existing solutions struggle with challenges related to security, scalability, efficiency, and real-time collaboration. This study seeks to enhance these efforts by proposing a novel decentralized solution that leverages the Ethereum blockchain and IPFS for secure, efficient, and resilient code repository hosting and governance. Our approach introduces a hybrid architecture that combines the immutable and decentralized nature of blockchain with the efficiency of IPFS for off-chain storage. To facilitate real-time collaboration, we integrate a temporary centralized Middleman IPFS that manages transaction processing and enhances operational efficiency without compromising long-term security. This Middleman IPFS acts as an intermediary, balancing the speed of centralized systems with the resilience of decentralized architectures. Our system uses smart contracts to maintain access control and key management by dynamically verifying access rights, ensuring that only authorized users can retrieve and decrypt data stored on IPFS. This integration allows for secure, real-time collaboration in environments where multiple collaborators need concurrent access to shared resources. Our system employs a hybrid encryption scheme that combines symmetric and asymmetric cryptography. The encrypted keys are stored on the blockchain, while IPFS handles the efficient storage of the codebase itself, with a Middleman IPFS maintaining concurrent collaboration, providing a robust and scalable solution for managing large-scale, collaborative coding projects., Comment: 24 pages, 3 figures, submitted manuscript to Plos one journal

The exponential growth of the Internet of Things (IoT) has significantly increased the complexity and volume of cybersecurity threats, necessitating the development of advanced, scalable, and interpretable security frameworks. This paper presents an innovative, comprehensive framework for real-time IoT attack detection and response that leverages Machine Learning (ML), Explainable AI (XAI), and Large Language Models (LLM). By integrating XAI techniques such as SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) with a model-independent architecture, we ensure our framework's adaptability across various ML algorithms. Additionally, the incorporation of LLMs enhances the interpretability and accessibility of detection decisions, providing system administrators with actionable, human-understandable explanations of detected threats. Our end-to-end framework not only facilitates a seamless transition from model development to deployment but also represents a real-world application capability that is often lacking in existing research. Based on our experiments with the CIC-IOT-2023 dataset \cite{neto2023ciciot2023}, Gemini and OPENAI LLMS demonstrate unique strengths in attack mitigation: Gemini offers precise, focused strategies, while OPENAI provides extensive, in-depth security measures. Incorporating SHAP and LIME algorithms within XAI provides comprehensive insights into attack detection, emphasizing opportunities for model improvement through detailed feature analysis, fine-tuning, and the adaptation of misclassifications to enhance accuracy., Comment: 6 pages, 1 figure, Accepted in 2024 IEEE WF-IoT Conference

Effective internet traffic prediction in smaller ISP networks is challenged by limited data availability. This paper explores this issue using transfer learning and data augmentation techniques with two LSTM-based models, LSTMSeq2Seq and LSTMSeq2SeqAtn, initially trained on a comprehensive dataset provided by Juniper Networks and subsequently applied to smaller datasets. The datasets represent real internet traffic telemetry, offering insights into diverse traffic patterns across different network domains. Our study revealed that while both models performed well in single-step predictions, multi-step forecasts were challenging, particularly in terms of long-term accuracy. In smaller datasets, LSTMSeq2Seq generally outperformed LSTMSeq2SeqAtn, indicating that higher model complexity does not necessarily translate to better performance. The models' effectiveness varied across different network domains, reflecting the influence of distinct traffic characteristics. To address data scarcity, Discrete Wavelet Transform was used for data augmentation, leading to significant improvements in model performance, especially in shorter-term forecasts. Our analysis showed that data augmentation is crucial in scenarios with limited data. Additionally, the study included an analysis of the models' variability and consistency, with attention mechanisms in LSTMSeq2SeqAtn providing better short-term forecasting consistency but greater variability in longer forecasts. The results highlight the benefits and limitations of different modeling approaches in traffic prediction. Overall, this research underscores the importance of transfer learning and data augmentation in enhancing the accuracy of traffic prediction models, particularly in smaller ISP networks with limited data availability., Comment: 16 pages, 7 Figures, Submitted to Elsevier Journal of Computer Communication

Objective: Variation of forearm orientation is one of the crucial factors that drastically degrades the forearm orientation invariant hand gesture recognition performance or the degree of freedom and limits the successful commercialization of myoelectric prosthetic hand or electromyogram (EMG) signal-based human-computer interfacing devices. This study investigates the impact of surface EMG electrode positions (elbow and forearm) on forearm orientation invariant hand gesture recognition. Methods: The study has been performed over 19 intact limbed subjects, considering 12 daily living hand gestures. The quality of the EMG signal is confirmed in terms of three indices. Then, the recognition performance is evaluated and validated by considering three training strategies, six feature extraction methods, and three classifiers. Results: The forearm electrode position provides comparable to or better EMG signal quality considering three indices. In this research, the forearm electrode position achieves up to 5.35% improved forearm orientation invariant hand gesture recognition performance compared to the elbow electrode position. The obtained performance is validated by considering six feature extraction methods, three classifiers, and real-time experiments. In addition, the forearm electrode position shows its robustness with the existence of recent works, considering recognition performance, investigated gestures, the number of channels, the dimensionality of feature space, and the number of subjects. Conclusion: The forearm electrode position can be the best choice for getting improved forearm orientation invariant hand gesture recognition performance. Significance: The performance of myoelectric prosthesis and human-computer interfacing devices can be improved with this optimized electrode position., Comment: 10 pages, 4 figures, 5 tables

Audio-driven 3D facial animation synthesis has been an active field of research with attention from both academia and industry. While there are promising results in this area, recent approaches largely focus on lip-sync and identity control, neglecting the role of emotions and emotion control in the generative process. That is mainly due to the lack of emotionally rich facial animation data and algorithms that can synthesize speech animations with emotional expressions at the same time. In addition, majority of the models are deterministic, meaning given the same audio input, they produce the same output motion. We argue that emotions and non-determinism are crucial to generate diverse and emotionally-rich facial animations. In this paper, we propose ProbTalk3D a non-deterministic neural network approach for emotion controllable speech-driven 3D facial animation synthesis using a two-stage VQ-VAE model and an emotionally rich facial animation dataset 3DMEAD. We provide an extensive comparative analysis of our model against the recent 3D facial animation synthesis approaches, by evaluating the results objectively, qualitatively, and with a perceptual user study. We highlight several objective metrics that are more suitable for evaluating stochastic outputs and use both in-the-wild and ground truth data for subjective evaluation. To our knowledge, that is the first non-deterministic 3D facial animation synthesis method incorporating a rich emotion dataset and emotion control with emotion labels and intensity levels. Our evaluation demonstrates that the proposed model achieves superior performance compared to state-of-the-art emotion-controlled, deterministic and non-deterministic models. We recommend watching the supplementary video for quality judgement. The entire codebase is publicly available (https://github.com/uuembodiedsocialai/ProbTalk3D/)., Comment: 14 pages, 9 figures, 3 tables. Includes code. Accepted at ACM SIGGRAPH MIG 2024

The Columbia Stellarator eXperiment (CSX), currently being designed at Columbia University, aims to test theoretical predictions related to QA plasma behavior, and to pioneer the construction of an optimized stellarator using three-dimensional, non-insulated high-temperature superconducting (NI-HTS) coils. The magnetic configuration is generated by a combination of two circular planar poloidal field (PF) coils and two 3D-shaped interlinked (IL) coils, with the possibility to add windowpane coils to enhance shaping and experimental flexibility. The PF coils and vacuum vessel are repurposed from the former Columbia Non-Neutral Torus (CNT) experiment, while the IL coils will be custom-wound in-house using NI-HTS tapes. To obtain a plasma shape that meets the physics objectives with a limited number of coils, novel single-stage optimization techniques are employed, optimizing both the plasma and coils concurrently, in particular targeting a tight aspect ratio QA plasma and minimized strain on the HTS tape. Despite the increased complexity due to the expanded degrees of freedom, these methods successfully identify optimized plasma geometries that can be realized by coils meeting engineering specifications. This paper discusses the derivation of the constraints and objectives specific to CSX, and describe how two recently developed single-stage optimization methodologies are applied to the design of CSX. A set of selected configurations for CSX is then described in detail., Comment: 36 pages, 12 figures

The shift toward more renewable energy sources and distributed generation in smart grids has underscored the significance of modeling and analyzing modern power systems as cyber-physical systems (CPS). This transformation has highlighted the importance of cyber and cyber-physical properties of modern power systems for their reliable operation. Graph theory emerges as a pivotal tool for understanding the complex interactions within these systems, providing a framework for representation and analysis. The challenge is vetting these graph theoretic methods and other estimates of system behavior from mathematical models against reality. High-fidelity emulation and/or simulation can help answer this question, but the comparisons have been understudied. This paper employs graph-theoretic metrics to assess node risk and criticality in three distinct case studies, using a Python-based discrete-event simulation called SimPy. Results for each case study show that combining graph theory and simulation provides a topology-informed security assessment. These tools allow us to identify critical network nodes and evaluate their performance and reliability under a cyber threat such as denial of service threats., Comment: 2024 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)

Ultra-low mass primordial black holes (PBH), briefly dominating the expansion of the universe, would leave detectable imprints in the secondary stochastic gravitational wave background (SGWB). Such a scenario leads to a characteristic doubly peaked spectrum of SGWB and strongly depends on the Hawking evaporation of such light PBHs. However, these observable signatures are significantly altered if the memory burden effect during the evaporation of PBHs is taken into account. We show that for the SGWB induced by PBH density fluctuations, the memory burden effects on the Hawking evaporation of ultra-low mass PBHs can mimic the signal arising due to the non-standard reheating epoch before PBH domination. Finally, we point out that this degeneracy can be broken by the simultaneous detection of the first peak in the SGWB, which is typically induced by the inflationary adiabatic perturbations., Comment: v1: 27 pages, 4 figures; v2: 28 pages, 4 figures, accepted for publication in JHEP

The black-box nature of Convolutional Neural Networks (CNNs) and their reliance on large datasets limit their use in complex domains with limited labeled data. Physics-Guided Neural Networks (PGNNs) have emerged to address these limitations by integrating scientific principles and real-world knowledge, enhancing model interpretability and efficiency. This paper proposes a novel Physics-Guided CNN (PGCNN) architecture that incorporates dynamic, trainable, and automated LLM-generated, widely recognized rules integrated into the model as custom layers to address challenges like limited data and low confidence scores. The PGCNN is evaluated on multiple datasets, demonstrating superior performance compared to a baseline CNN model. Key improvements include a significant reduction in false positives and enhanced confidence scores for true detection. The results highlight the potential of PGCNNs to improve CNN performance for broader application areas.

This work provides an optimization mechanism to ensure the compatibility of ReBCO (Rare-earth Barium Copper Oxide) high-temperature superconducting (HTS) tapes with non-planar stellarator coils. ReBCO coils enable higher field strengths and operating temperatures for the magnet systems of future fusion reactors but are sensitive to strain due to their brittle, ceramic functional layer. We have implemented a metric to optimize strain on stellarator coils made from ReBCO superconductors into the stellarator optimization framework \texttt{SIMSOPT} and used it to design new stellarator coil configurations. To ensure structural integrity of coils wound with HTS tape, we introduce a penalty on binormal curvature and torsion along a coil. It can be used to optimize the orientation of the winding path for a given coil filament or to jointly optimize orientation and coil filament. We apply the strain optimization to three cases. For the EPOS (Electrons and Positrons in an Optimized Stellarator) design, we combine the strain penalty with an objective for quasisymmetry into a single-stage optimization; this enables us to find a configuration with excellent quasisymmetry at the smallest possible size compatible with the use of ReBCO tape. For CSX (Columbia Stellarator eXperiment), in addition to HTS strain, we add a penalty to prevent net tape rotation to ease the coil winding process. If the strain is calculated for a coil at reactor scale, we find a considerable variation of the binormal and torsional strain over the cross section of the large winding pack (0.5\,m x 0.5\,m). By exploiting the overall orientation of the winding pack as a degree of freedom, we can reduce binormal and torsional strains below limits for every ReBCO stack., Comment: 14 pages, 14 figures

Stream processing applications have been widely adopted due to real-time data analytics demands, e.g., fraud detection, video analytics, IoT applications. Unfortunately, prototyping and testing these applications is still a cumbersome process for developers that usually requires an expensive testbed and deep multi-disciplinary expertise, including in areas such as networking, distributed systems, and data engineering. As a result, it takes a long time to deploy stream processing applications into production and yet users face several correctness and performance issues. In this paper, we present stream2gym, a tool for the fast prototyping of large-scale distributed stream processing applications. stream2gym builds on Mininet, a widely adopted network emulation platform, and provides a high-level interface to enable developers to easily test their applications under various operating conditions. We demonstrate the benefits of stream2gym by prototyping and testing several applications as well as reproducing key findings from prior research work in video analytics and network traffic monitoring. Moreover, we show stream2gym presents accurate results compared to a hardware testbed while consuming a small amount of resources (enough to be supported in a single commodity laptop even when emulating a dozen of processing nodes)., Comment: 11 pages, 9 figures (excluding subfigures), accepted in IEEE ICDCS 2023

Plant diseases significantly impact agricultural productivity, resulting in economic losses and food insecurity. Prompt and accurate detection is crucial for the efficient management and mitigation of plant diseases. This study investigates advanced techniques in plant disease detection, emphasizing the integration of image processing, machine learning, deep learning methods, and mobile technologies. High-resolution images of plant leaves were captured and analyzed using convolutional neural networks (CNNs) to detect symptoms of various diseases, such as blight, mildew, and rust. This study explores 14 classes of plants and diagnoses 26 unique plant diseases. We focus on common diseases affecting various crops. The model was trained on a diverse dataset encompassing multiple crops and disease types, achieving 98.14% accuracy in disease diagnosis. Finally integrated this model into mobile apps for real-time disease diagnosis.

The recent surge in AI-generated songs presents exciting possibilities and challenges. While these inventions democratize music creation, they also necessitate the ability to distinguish between human-composed and synthetic songs to safeguard artistic integrity and protect human musical artistry. Existing research and datasets in fake song detection only focus on singing voice deepfake detection (SVDD), where the vocals are AI-generated but the instrumental music is sourced from real songs. However, these approaches are inadequate for detecting contemporary end-to-end artificial songs where all components (vocals, music, lyrics, and style) could be AI-generated. Additionally, existing datasets lack music-lyrics diversity, long-duration songs, and open-access fake songs. To address these gaps, we introduce SONICS, a novel dataset for end-to-end Synthetic Song Detection (SSD), comprising over 97k songs (4,751 hours) with over 49k synthetic songs from popular platforms like Suno and Udio. Furthermore, we highlight the importance of modeling long-range temporal dependencies in songs for effective authenticity detection, an aspect entirely overlooked in existing methods. To utilize long-range patterns, we introduce SpecTTTra, a novel architecture that significantly improves time and memory efficiency over conventional CNN and Transformer-based models. In particular, for long audio samples, our top-performing variant outperforms ViT by 8% F1 score while being 38% faster and using 26% less memory. Additionally, in comparison with ConvNeXt, our model achieves 1% gain in F1 score with 20% boost in speed and 67% reduction in memory usage. Other variants of our model family provide even better speed and memory efficiency with competitive performance., Comment: Updated with correction

We have discussed in detail how neutrinos produced from inflaton solely through gravitational interaction can successfully reheat the universe. For this, we have introduced the well-known Type-I seesaw neutrino model. Depending on seesaw model parameters, two distinct reheating histories have been realized and dubbed as i) Neutrino dominating: Following the inflaton domination, the universe becomes neutrino dominated, and their subsequent decay concludes the reheating process, and ii) Neutrino heating: Despite being sub-dominant compared to inflaton energy, neutrinos efficiently heat the thermal bath and produce the radiation dominated universe. Imposing baryon asymmetric yield, the $\Delta N_{\rm eff}$ constraint at Big Bang Nucleosynthesis (BBN) considering primordial gravitational waves (PGW), we have arrived at the following constraints on reheating equation of state to lie within $0.5\lesssim w_\phi\lesssim1.0$. In these neutrino-driven reheating backgrounds, we further performed a detailed analysis of both thermal and non-thermal production of dark matter (DM), invoking two minimal models, namely the Higgs portal DM and classical QCD pseudo scalar axion. An interesting correlation between seemingly uncorrelated DM and Type-I seesaw parameters has emerged when confronting various direct and indirect observations. When DMs are set to freeze-in, freeze-out, or oscillate during reheating, new parameter spaces open, which could be potentially detectable in future experiments, paving an indirect way to look into the early universe in the laboratory., Comment: 24 pages,13 figures, 3 tables

Low-noise amplifiers are of great importance in the field of quantum technologies. We study a thermally driven parametric amplifier based on a superconductor-insulator-graphene-insulator-superconductor (SIGIS) junction coupled to a superconducting microwave cavity. The strong non-linearity in the temperature dependence of our device leads to thermal self-modulation that produces impedance oscillations at frequencies around twice the angular cavity resonance frequency $\omega_\mathrm{r}$. In particular, reactance modulation of the effective capacitance yields a gain of 18.6 dB over a frequency span of 125 kHz with a minimum noise temperature of $T_\mathrm{N} = 1.4$ K. Our theoretical modelling gives insight into the exact mixing processes, confirmation of the electron-phonon coupling parameter and possible improvements of the studied system., Comment: 9 pages, 5 figures, v2 changes to fig 2 and fig.2 caption

Aligning large language models (LLMs) with a human reasoning approach ensures that LLMs produce morally correct and human-like decisions. Ethical concerns are raised because current models are prone to generating false positives and providing malicious responses. To contribute to this issue, we have curated an ethics dataset named Dataset for Aligning Reasons (DFAR), designed to aid in aligning language models to generate human-like reasons. The dataset comprises statements with ethical-unethical labels and their corresponding reasons. In this study, we employed a unique and novel fine-tuning approach that utilizes ethics labels and their corresponding reasons (L+R), in contrast to the existing fine-tuning approach that only uses labels (L). The original pre-trained versions, the existing fine-tuned versions, and our proposed fine-tuned versions of LLMs were then evaluated on an ethical-unethical classification task and a reason-generation task. Our proposed fine-tuning strategy notably outperforms the others in both tasks, achieving significantly higher accuracy scores in the classification task and lower misalignment rates in the reason-generation task. The increase in classification accuracies and decrease in misalignment rates indicate that the L+R fine-tuned models align more with human ethics. Hence, this study illustrates that injecting reasons has substantially improved the alignment of LLMs, resulting in more human-like responses. We have made the DFAR dataset and corresponding codes publicly available at https://github.com/apurba-nsu-rnd-lab/DFAR., Comment: Accepted in ICPR 2024