arXiv subjects

Nishant Kumar

Publications and source records attributed to Nishant Kumar.

At least 19 recordsLinked to original sources

arXiv2026

Toward a Stable and Deployable Adaptive Chirplet Transform: Residual Projection, Hybrid GPU Acceleration, and Multi-Channel Scalability

The Adaptive Chirplet Transform is a flexible framework that can decompose non-stationary signals into sparse chirplets; it has been applied to signals such as electroencephalography, electromyography and radar. However, the practical deployment of this transform has been hindered by two challenges: algorithmic instability in prior implementations, which can lead to divergent decompositions, and the computational cost of searching over a high-dimensional parameter space. This paper addresses both by a sequence of contributions. Firstly, unit normalization and residual-based projection are introduced to align the decomposition with Matching Pursuit Theory, thereby eliminating divergence and substantially reducing residual error across all signal domains, as demonstrated on three representative signal types. A hybrid CPU-GPU architecture offloads chirplet family generation to the CPU while parallelizing the search on the GPU, removing bottlenecks in CPU-only search and GPU-only generation, achieving speedups of 6.6-7.38x on desktop hardware, with consistent gains observed across laptop and embedded platforms. Multichannel batching enabled simultaneous multi-signal processing, amplifying the speedup, which scaled from 3.94x for a single channel to 8.22x at 10 channels. Finally, a hierarchical coarse-to-fine search, inspired by Logon Expectation Maximization, is introduced. This reduced peak memory usage below 1 GB while maintaining similar reconstruction quality, at the cost of longer runtime. Together, these contributions establish a correct, stable and practically deployable foundation for chirplet-based signal decomposition. Index Terms: Chirplet Transform, GPU Computing, Matching Pursuit, Signal Decomposition, Sparse Representation, Time-Frequency Analysis

eess.SP↗

arXiv2026

Zero Collapse: A Failure Mode of Policy Gradient Methods in Discontinuous Reward Environments

Bidding in repeated auctions is a central challenge for reinforcement learning (RL), combining continuous control with the strategic complexities of digital advertising. While policy gradient and value-based methods seem well-suited for these settings, they often struggle with the discontinuous, "cliff-like" nature of auction reward landscapes. In a first-price auction, for example, a bidder receives zero reward until they cross a specific threshold, after which the reward decreases as the bid increases. This creates a landscape of flat, zero-reward regions separated by sharp boundaries. We identify a fundamental failure mode in this setting termed "zero collapse." We show that stochastic exploration and gradient-based updates can cause policies to overshoot optimal high-reward regions and enter flat, zero-reward regimes. Once there, the lack of an informative gradient signal makes recovery extremely sample-inefficient, effectively trapping the agent. We find that actor-critic methods are particularly susceptible, as biased value estimates can accelerate this movement toward unstable regions. Our contributions include: (1) a mechanistic explanation of how discontinuous rewards lead to vanishing signals and zero collapse; (2) an analysis of the interaction between policy stochasticity and step size; and (3) an empirical demonstration of this phenomenon across REINFORCE and actor-critic variants. We propose practical mitigation strategies involving initialization and architectural choices to improve stability. Finally, we introduce a formal RL framework for auction environments highlighting their unique structural properties.

cs.LG↗

arXiv2026

Oxygen-Pressure-Limited Recovery of the Hematite {\alpha}-Fe$_2$O$_3$(0001) Surface from a Reduced Fe$_3$O$_4$(111)-Like Layer

The oxidation kinetics of hematite {\alpha}-Fe$_2$O$_3$(0001) surfaces are vital for its applications in catalysis, environmental remediation, and industrial processes. Despite prior studies, the roles of temperature, oxygen partial pressure, and oxygen chemical potential in controlling nucleation and growth kinetics are not fully understood. Using real-time Low Energy Electron Microscopy/Diffraction (LEEM/LEED), we systematically investigate the oxidation of a reduced Fe$_3$O$_4$(111)-like surface layer to hematite under controlled conditions. We show that complete recovery of the hematite surface termination is closely linked to the nucleation and lateral growth of a two-dimensional honeycomb (H) phase. While higher temperatures accelerate nucleation, they slow lateral growth at constant oxygen pressure, indicating that oxygen supply limits the oxidation rate. Below an oxygen partial pressure threshold (~2$\times$10$^{-6}$ mbar), growth dramatically slows, underscoring the critical role of oxygen availability. Below a certain oxygen pressure threshold, the growth time rapidly increases. Our study elucidates the interplay between thermodynamics and kinetics in hematite surface oxidation, informing strategies to optimize surface properties for catalytic and industrial processes.

cond-mat.mtrl-sci↗

arXiv2026

End-to-end data-driven prediction of urban airflow and pollutant dispersion

Climate change and the rapid growth of urban populations are intensifying environmental stresses within cities, making the behavior of urban atmospheric flows a critical factor in public health, energy use, and overall livability. This study targets to develop fast and accurate models of urban pollutant dispersion to support decision-makers, enabling them to implement mitigation measures in a timely and cost-effective manner. To reach this goal, an end-to-end data-driven approach is proposed to model and predict the airflow and pollutant dispersion in a street canyon in skimming flow regime. A series of time-resolved snapshots obtained from large eddy simulation (LES) serves as the database. The proposed framework is based on four fundamental steps. Firstly, a reduced basis is obtained by spectral proper orthogonal decomposition (SPOD) of the database. The projection of the time series snapshot data onto the SPOD modes (time-domain approach) provides the temporal coefficients of the dynamics. Secondly, a nonlinear compression of the temporal coefficients is performed by autoencoder to reduce further the dimensionality of the problem. Thirdly, a reduced-order model (ROM) is learned in the latent space using Long Short-Term Memory (LSTM) netowrks. Finally, the pollutant dispersion is estimated from the predicted velocity field through convolutional neural network that maps both fields. The results demonstrate the efficacy of the model in predicting the instantaneous as well as statistically stationary fields over long time horizon.

cs.LG↗

arXiv2025

Assessment of East-West (E-W) and South-North (S-N) facing Vertical Bifacial Photovoltaic Modules for Agrivoltaics and Dual-Land Use Applications in India

Deploying vertical bifacial PV modules can play a significant role in agrivoltaics, fencing walls, noise barriers, building integrated photovoltaics (BIPV), solar PV for electric vehicles, and many other applications. This research work presents the performance comparison of vertical bifacial photovoltaic (VBPV) modules facing East-West (E-W) and South-North (S-N) directions. Also, the VBPV modules are compared with vertical and tilted south-facing monofacial PV modules. Six PV modules (monofacial and bifacial) were installed at the rooftop of IIT Bhilai academic building, Raipur (21.16{\deg} N, 81.65{\deg} E), India, and studied for a year from May 2022 to April 2023. The results show that the E-W facing VBPV module gives two production peaks, one in the morning and another in the evening, as compared to the single notable rise at midday observed for a monofacial module. From a series of experiments, 19 days of data were collected over the one-year period from May 2022 to April 2023, with specific inclusion of important days like solstices and equinoxes. In addition, the energy generation results are compared with PVsyst simulations, while also addressing the limitations of the PVsyst simulation of vertical PV modules. E-W bifacial generation is higher than S-N bifacial and south-facing monofacial modules from February to April. The VBPV modules in E-W and S-N orientations present a promising opportunity for expanding the agrivoltaics sector in tropical and sub-tropical countries, like India. This has huge implications for addressing the sustainable development goals by simultaneously contributing to sustainable land management, green energy generation, energy security and water conservation in the vast geo-climatic expanse of tropics.

eess.SY↗

arXiv2025

Let's Play Across Cultures: A Large Multilingual, Multicultural Benchmark for Assessing Language Models' Understanding of Sports

Language Models (LMs) are primarily evaluated on globally popular sports, often overlooking regional and indigenous sporting traditions. To address this gap, we introduce \textbf{\textit{CultSportQA}}, a benchmark designed to assess LMs' understanding of traditional sports across 60 countries and 6 continents, encompassing four distinct cultural categories. The dataset features 33,000 multiple-choice questions (MCQs) across text and image modalities, each of which is categorized into three key types: history-based, rule-based, and scenario-based. To evaluate model performance, we employ zero-shot, few-shot, and chain-of-thought (CoT) prompting across a diverse set of Large Language Models (LLMs), Small Language Models (SLMs), and Multimodal Large Language Models (MLMs). By providing a comprehensive multilingual and multicultural sports benchmark, \textbf{\textit{CultSportQA}} establishes a new standard for assessing AI's ability to understand and reason about traditional sports.

cs.CL↗

arXiv2025

Toward Quantum Enabled Solutions for Real-Time Currency Arbitrage in Financial Markets

Currency arbitrage leverages price discrepancies in currency exchange rates across different currency pairs to gain risk-free profits. It involves multiple trading, where short-lived price discrepancies require real-time, high-speed processing of vast solution space, posing challenges for classical computing. In this work, we formulate an enhanced mathematical model for the currency arbitrage problem by adding simple cycle preservation constraints, which guarantee trading cycle validity and eliminate redundant or infeasible substructures. To solve this model, we use and benchmark various solvers, including Quantum Annealing (QA), gate-based quantum approaches such as Variational Quantum Algorithm with Adaptive Cost Encoding (ACE), as well as classical solvers such as Gurobi and classical meta heuristics such as Tabu Search (TS). We propose a classical multi-bit swap post-processing to improve the solution generated by ACE. Using real-world currency exchange data, we compare these methods in terms of both arbitrage profit and execution time, the two key performance metrics. Our results give insight into the current capabilities and limitations of quantum methods for real-time financial use cases.

quant-ph↗

arXiv2025

Generation of Indian Sign Language Letters, Numbers, and Words

Sign language, which contains hand movements, facial expressions and bodily gestures, is a significant medium for communicating with hard-of-hearing people. A well-trained sign language community communicates easily, but those who don't know sign language face significant challenges. Recognition and generation are basic communication methods between hearing and hard-of-hearing individuals. Despite progress in recognition, sign language generation still needs to be explored. The Progressive Growing of Generative Adversarial Network (ProGAN) excels at producing high-quality images, while the Self-Attention Generative Adversarial Network (SAGAN) generates feature-rich images at medium resolutions. Balancing resolution and detail is crucial for sign language image generation. We are developing a Generative Adversarial Network (GAN) variant that combines both models to generate feature-rich, high-resolution, and class-conditional sign language images. Our modified Attention-based model generates high-quality images of Indian Sign Language letters, numbers, and words, outperforming the traditional ProGAN in Inception Score (IS) and Fr\'echet Inception Distance (FID), with improvements of 3.2 and 30.12, respectively. Additionally, we are publishing a large dataset incorporating high-quality images of Indian Sign Language alphabets, numbers, and 129 words.

cs.CV↗

arXiv2025

CBTOPE2: An improved method for predicting of conformational B-cell epitopes in an antigen from its primary sequence

In 2009, our group pioneered a novel method CBTOPE for predicting conformational B-cell epitopes in a protein from its amino acid sequence, which received extensive citations from the scientific community. In a recent study, Cia et al. (2023) evaluated the performance of conformational B-cell epitope prediction methods on a well-curated dataset, revealing that most approaches, including CBTOPE, exhibited poor performance. One plausible cause of this diminished performance is that available methods were trained on datasets that are both limited in size and outdated in content. In this study, we present an enhanced version of CBTOPE, trained, tested, and evaluated using the well-curated dataset from Cai et al. (2023). Initially, we developed machine learning-based models using binary profiles, achieving a maximum AUC of 0.58 on the validation dataset. The performance of our method improved significantly from an AUC of 0.58 to 0.63 when incorporating evolutionary information in the form of a Position-Specific Scoring Matrix (PSSM) profile. Furthermore, the performance increased from an AUC of 0.63 to 0.64 when we integrated both the PSSM profile and relative solvent accessibility (RSA). All models were trained, tested, and optimized on the training dataset using five-fold cross-validation. The final performance of our models was assessed using a validation or independent dataset that was not used during hyperparameter optimization. To facilitate scientific community working in the field of subunit vaccine, we develop a standalone software and web server CBTOPE2 (https://webs.iiitd.edu.in/raghava/cbtope2/).

q-bio.BM↗

arXiv2025

In silico tool for identification of colorectal cancer from cell-free DNA biomarkers

Colorectal cancer remains a major global health concern, with early detection being pivotal for improving patient outcomes. In this study, we leveraged high throughput methylation profiling of cellfree DNA to identify and validate diagnostic biomarkers for CRC. The GSE124600 study data were downloaded from the Gene Expression Omnibus, as the discovery cohort, comprising 142 CRC and 132 normal cfDNA methylation profiles obtained via MCTA seq. After preprocessing and filtering, 97,863 CpG sites were retained for further analysis. Differential methylation analysis using statistical tests identified 30,791 CpG sites as significantly altered in CRC samples, where p is less than 0.05. Univariate scoring enabled the selection of top ranking features, which were further refined using multiple feature selection algorithms, including Recursive Feature Elimination, Sequential Feature Selection, and SVC L1. Various machine learning models such as Logistic Regression, Support Vector Machines, Random Forest, and Multi layer Perceptron were trained and tested using independent validation datasets. The best performance was achieved with an MLP model trained on 25 features selected by RFE, reaching an AUROC of 0.89 and MCC of 0.78 on validation data. Additionally, a deep learning based convolutional neural network achieved an AUROC of 0.78. Functional annotation of the most predictive CpG sites identified several genes involved in key cellular processes, some of which were validated for differential expression in CRC using the GEPIA2 platform. Our study highlights the potential of cfDNA methylation markers combined with ML and DL models for noninvasive and accurate CRC detection, paving the way for clinically relevant diagnostic tools.

q-bio.GN↗

arXiv2025

MAP Format for Representing Chemical Modifications, Annotations, and Mutations in Protein Sequences: An Extension of the FASTA Format

Several formats, including FASTA, PIR, GenBank, EMBL, and GCG, have been developed for representing protein sequences composed of natural amino acids. Among these, FASTA remains the most widely used due to its simplicity and human readability. However, FASTA lacks the capability to represent chemically modified or non-natural residues, as well as structural annotations and mutations in protein variants. To address some of these limitations, the PEFF format was recently introduced as an extension of FASTA. Additionally, formats such as HELM and BILN have been proposed to represent amino acids and their modifications at the atomic level. Despite their advancements, these formats have not achieved widespread adoption within the bioinformatics community due to their complexity. To complement existing formats and overcome current challenges, we propose a new format called MAP (Modification and Annotation in Proteins), which enables comprehensive annotation of protein sequences. MAP introduces meta tags in the header for protein-level annotations and inline tags within the sequence for residue-level modifications. In this format, standard one-letter amino acid codes are augmented with curly-brace tags to denote various modifications, including phosphorylation, acetylation, non-natural residues, cyclization, and other residue-specific features. The header metadata also captures information such as organism, function, and sequence variants. We describe the structure, objectives, and capabilities of the MAP format and demonstrate its application in bioinformatics, particularly in the domain of protein therapeutics. To facilitate community adoption, we are developing a comprehensive suite of MAP-format resources, including a detailed manual, annotated datasets, and conversion tools, available at http://webs.iiitd.edu.in/raghava/maprepo/.

q-bio.BM↗

arXiv2025

Single Ultrabright Fluorescent Silica Nanoparticles Can Be Used as Individual Fast Real-Time Nanothermometers

Optical-based nanothermometry represents a transformative approach for precise temperature measurements at the nanoscale, which finds versatile applications across biology, medicine, and electronics. The assembly of ratiometric fluorescent 40 nm nanoparticles designed to serve as individual nanothermometers is introduced here. These nanoparticles exhibit unprecedented sensitivity (11% /K) and temperature resolution 128 \, \mathrm{K} \cdot \mathrm{Hz}^{-1/2} \cdot \mathrm{W} \cdot \mathrm{cm}^{-2}, outperforming existing optical nanothermometers by factors of 2-6 and 455, respectively. The enhanced performance is attributed to the encapsulation of fluorescent molecules with high density inside the mesoporous matrix. It becomes possible after incorporating hydrophobic groups into the silica matrix, which effectively prevents water ingress and dye leaking. A practical application of these nanothermometers is demonstrated using confocal microscopy, showcasing their ability to map temperature distributions accurately. This methodology is compatible with any fluorescent microscope capable of recording dual fluorescent channels in any transparent medium or on a sample surface. This work not only sets a new benchmark for optical nano-thermometry but also provides a relatively simple yet powerful tool for exploring thermal phenomena at the nanoscale across various scientific domains.

physics.optics↗

arXiv2024

FusionINN: Decomposable Image Fusion for Brain Tumor Monitoring

Image fusion typically employs non-invertible neural networks to merge multiple source images into a single fused image. However, for clinical experts, solely relying on fused images may be insufficient for making diagnostic decisions, as the fusion mechanism blends features from source images, thereby making it difficult to interpret the underlying tumor pathology. We introduce FusionINN, a novel decomposable image fusion framework, capable of efficiently generating fused images and also decomposing them back to the source images. FusionINN is designed to be bijective by including a latent image alongside the fused image, while ensuring minimal transfer of information from the source images to the latent representation. To the best of our knowledge, we are the first to investigate the decomposability of fused images, which is particularly crucial for life-sensitive applications such as medical image fusion compared to other tasks like multi-focus or multi-exposure image fusion. Our extensive experimentation validates FusionINN over existing discriminative and generative fusion methods, both subjectively and objectively. Moreover, compared to a recent denoising diffusion-based fusion model, our approach offers faster and qualitatively better fusion results.

eess.IV↗

arXiv2023

Quantile-based Maximum Likelihood Training for Outlier Detection

Discriminative learning effectively predicts true object class for image classification. However, it often results in false positives for outliers, posing critical concerns in applications like autonomous driving and video surveillance systems. Previous attempts to address this challenge involved training image classifiers through contrastive learning using actual outlier data or synthesizing outliers for self-supervised learning. Furthermore, unsupervised generative modeling of inliers in pixel space has shown limited success for outlier detection. In this work, we introduce a quantile-based maximum likelihood objective for learning the inlier distribution to improve the outlier separation during inference. Our approach fits a normalizing flow to pre-trained discriminative features and detects the outliers according to the evaluated log-likelihood. The experimental evaluation demonstrates the effectiveness of our method as it surpasses the performance of the state-of-the-art unsupervised methods for outlier detection. The results are also competitive compared with a recent self-supervised approach for outlier detection. Our work allows to reduce dependency on well-sampled negative training data, which is especially important for domains like medical diagnostics or remote sensing.

cs.CV↗

arXiv2023

Uncertainty Quantification for Image-based Traffic Prediction across Cities

Despite the strong predictive performance of deep learning models for traffic prediction, their widespread deployment in real-world intelligent transportation systems has been restrained by a lack of interpretability. Uncertainty quantification (UQ) methods provide an approach to induce probabilistic reasoning, improve decision-making and enhance model deployment potential. To gain a comprehensive picture of the usefulness of existing UQ methods for traffic prediction and the relation between obtained uncertainties and city-wide traffic dynamics, we investigate their application to a large-scale image-based traffic dataset spanning multiple cities and time periods. We compare two epistemic and two aleatoric UQ methods on both temporal and spatio-temporal transfer tasks, and find that meaningful uncertainty estimates can be recovered. We further demonstrate how uncertainty estimates can be employed for unsupervised outlier detection on changes in city traffic dynamics. We find that our approach can capture both temporal and spatial effects on traffic behaviour in a representative case study for the city of Moscow. Our work presents a further step towards boosting uncertainty awareness in traffic prediction tasks, and aims to highlight the value contribution of UQ methods to a better understanding of city traffic dynamics.

cs.CV↗

arXiv2023

Learning to reconstruct the bubble distribution with conductivity maps using Invertible Neural Networks and Error Diffusion

Electrolysis is crucial for eco-friendly hydrogen production, but gas bubbles generated during the process hinder reactions, reduce cell efficiency, and increase energy consumption. Additionally, these gas bubbles cause changes in the conductivity inside the cell, resulting in corresponding variations in the induced magnetic field around the cell. Therefore, measuring these gas bubble-induced magnetic field fluctuations using external magnetic sensors and solving the inverse problem of Biot-Savart Law allows for estimating the conductivity in the cell and, thus, bubble size and location. However, determining high-resolution conductivity maps from only a few induced magnetic field measurements is an ill-posed inverse problem. To overcome this, we exploit Invertible Neural Networks (INNs) to reconstruct the conductivity field. Our qualitative results and quantitative evaluation using random error diffusion show that INN achieves far superior performance compared to Tikhonov regularization.

eess.IV↗

arXiv2023

Towards Computational Performance Engineering for Unsupervised Concept Drift Detection -- Complexities, Benchmarking, Performance Analysis

Concept drift detection is crucial for many AI systems to ensure the system's reliability. These systems often have to deal with large amounts of data or react in real-time. Thus, drift detectors must meet computational requirements or constraints with a comprehensive performance evaluation. However, so far, the focus of developing drift detectors is on inference quality, e.g. accuracy, but not on computational performance, such as runtime. Many of the previous works consider computational performance only as a secondary objective and do not have a benchmark for such evaluation. Hence, we propose and explain performance engineering for unsupervised concept drift detection that reflects on computational complexities, benchmarking, and performance analysis. We provide the computational complexities of existing unsupervised drift detectors and discuss why further computational performance investigations are required. Hence, we state and substantiate the aspects of a benchmark for unsupervised drift detection reflecting on inference quality and computational performance. Furthermore, we demonstrate performance analysis practices that have proven their effectiveness in High-Performance Computing, by tracing two drift detectors and displaying their performance data.

cs.LG↗

arXiv2023

Hematite ${\alpha}-Fe_{2}O_{3}(0001)$ in top and side view: resolving long-standing controversies about its surface structure

Hematite ${\alpha}-Fe_{2}O_{3}(0001)$ is the most-investigated iron oxide model system in photo and electrocatalytic research. The rich chemistry of Fe and O allows for many bulk and surface transformations, but their control is challenging. This has led to controversies regarding the structure of the topmost layers. This comprehensive study combines surface methods (nc-AFM, STM, LEED, and XPS) complemented by structural and chemical analysis of the near-surface bulk (HRTEM and EELS). The results show that a compact 2D layer constitutes the topmost surface of ${\alpha}-Fe_{2}O_{3}(0001)$; it is locally corrugated due to the mismatch with the bulk. Assessing the influence of naturally-occurring impurities shows that these can force the formation of surface phases that are not stable on pure samples. Impurities can also cause the formation of ill-defined inclusions in the subsurface and modify the oxidation phase diagram of hematite. The results provide a significant step forward in determining the hematite surface structure that is crucial for accurately modeling catalytic reactions. Combining surface and cross-sectional imaging provided the full view that is essential for understanding the evolution of the near-surface region of oxide surfaces under oxidative conditions.

cond-mat.mtrl-sci↗