arXiv subjects

Aditya Mishra

Publications and source records attributed to Aditya Mishra.

At least 19 recordsLinked to original sources

arXiv2026

Simple Features and Honest Calibration for Ambivalence and Hesitancy Recognition in Video

We address ambivalence and hesitancy (A/H) recognition in the ABAW 2026 BAH Challenge: given a short interview video, predict whether the person shows signs of A/H. Our system combines affect-specialised text, audio, and visual representations with a small set of readable linguistic hesitation cues, fused by a reliability gate we call Affective Marker Fusion (AMF), and finished with a simple AP-weighted ensemble at a fixed decision threshold. We also introduce \emph{ASR-erased time}: speech recognisers delete fillers and hesitation pauses from the transcript, but the chunk timestamps keep the time those events took, and sixteen features built from these gaps form the strongest and most independent non-verbal channel we measured (AP $0.718$, correlation $0.11$--$0.36$ with all other members). Across controlled experiments we find three things: cross-modal conflict design does not reliably help on BAH; language is by far the strongest channel while affect-specialised audio is a useful second; and calibration matters more than architecture. Fitting ensemble weights and a threshold on the small validation split overfits: it scores $0.741$ macro-F1 on validation but only $0.690$ on the untouched test set. AP-weighting at a fixed threshold instead reaches $\mathbf{0.731}$ on test.

cs.CV↗

arXiv2026

Beyond Benchmarks: Continuous Edge Inference for Fine-Grained Roadside Perception

Continuous AI inference on resource-constrained edge hardware introduces deployment effects that are largely invisible to conventional benchmark evaluation, including temporal instability in streaming video, thermal throttling under sustained load, and workload-dependent performance variability. We present Edge-TSR, a deployment-oriented continuous edge inference system for sustained roadside perception on the NVIDIA Jetson Orin Nano. Edge-TSR integrates detection, tracking, fine-grained classification, and a lightweight track-aware temporal stabilization mechanism that improves streaming inference consistency with negligible computational overhead. Our central finding is that benchmark-centric evaluation systematically overstates deployed edge inference performance. Across three state-of-the-art baselines, we observe consistent 20-30% relative degradation when transitioning from static-image evaluation to real-world streaming deployment. Edge-TSR addresses this gap through temporal inference stabilization, recovering up to 10.16% classification accuracy over per-frame inference baselines while maintaining sustained real-time performance under continuous operation. We evaluate the complete system under diverse real-world deployment conditions, jointly characterizing inference quality, latency, throughput, and thermal behavior during long-duration operation. A 55-minute vehicular deployment over a 26 km route demonstrates sustained operation at 16.18 FPS within safe thermal limits on a single embedded device without cloud offload. Our findings show that deployment-aware evaluation and temporal inference stabilization are necessary components of continuously operating edge AI systems intended for real-world sensing deployments. We release a sample annotated streaming video evaluation dataset and full system implementation to support reproducible deployment-centric evaluation.

cs.CV↗

arXiv2025

Benchmarking Nighttime Traffic Sign Recognition with Illumination-Adaptive Detection and Semantic Attribute Reasoning

Traffic signboards are vital for road safety and intelligent transportation systems. Yet, recognizing traffic signs at night remains underexplored due to the scarcity of real-world public datasets capturing low-light degradations and distractor classes. Existing benchmarks are predominantly daytime and do not reflect challenges such as headlight glare, motion blur, sensor noise, and vandalized or ambiguous signage. To address these gaps, we introduce INTSD, a large-scale nighttime traffic sign dataset collected across diverse regions of India. INTSD contains street-level images spanning 41 traffic signboard classes, multiple distractor categories, and varied lighting and weather conditions, designed to support both detection and fine-grained classification under nighttime scenarios. To benchmark INTSD, we conduct extensive evaluations using state-of-the-art detection and classification models under standardized protocols. Additionally, we present LENS-Net, a strong baseline that integrates an end-to-end adaptive illumination-aware detector with a multimodal classifier that fuses vision-language representations with soft semantic attribute reasoning over learnable shape and color embeddings. Experiments demonstrate that models trained exclusively on daytime data fail substantially under real nighttime conditions - a gap that is recovered once INTSD is introduced in training, even when controlling for data volume. These results validate INTSD as a complementary nighttime training resource and establish competitive baselines for future research. The code and dataset are publicly available.

cs.CV↗

arXiv2025

A Tensor-Based Compiler and a Runtime for Neuron-Level DNN Certifier Specifications

The uninterpretability of DNNs has led to the adoption of abstract interpretation-based certification as a practical means to establish trust in real-world systems that rely on DNNs. However, the current landscape supports only a limited set of certifiers, and developing new ones or modifying existing ones for different applications remains difficult. This is because the mathematical design of certifiers is expressed at the neuron level, while their implementations are optimized and executed at the tensor level. This mismatch creates a semantic gap between design and implementation, making manual bridging both complex and expertise-intensive -- requiring deep knowledge in formal methods, high-performance computing, etc. We propose a compiler framework that automatically translates neuron-level specifications of DNN certifiers into tensor-based, layer-level implementations. This is enabled by two key innovations: a novel stack-based intermediate representation (IR) and a shape analysis that infers the implicit tensor operations needed to simulate the neuron-level semantics. During lifting, the shape analysis creates tensors in the minimal shape required to perform the corresponding operations. The IR also enables domain-specific optimizations as rewrites. At runtime, the resulting tensor computations exhibit sparsity tied to the DNN architecture. This sparsity does not align well with existing formats. To address this, we introduce g-BCSR, a double-compression format that represents tensors as collections of blocks of varying sizes, each possibly internally sparse. Using our compiler and g-BCSR, we make it easy to develop new certifiers and analyze their utility across diverse DNNs. Despite its flexibility, the compiler achieves performance comparable to hand-optimized implementations.

cs.CL↗

arXiv2025

Deep reinforcement learning-based longitudinal control strategy for automated vehicles at signalised intersections

Developing an autonomous vehicle control strategy for signalised intersections (SI) is one of the challenging tasks due to its inherently complex decision-making process. This study proposes a Deep Reinforcement Learning (DRL) based longitudinal vehicle control strategy at SI. A comprehensive reward function has been formulated with a particular focus on (i) distance headway-based efficiency reward, (ii) decision-making criteria during amber light, and (iii) asymmetric acceleration/ deceleration response, along with the traditional safety and comfort criteria. This reward function has been incorporated with two popular DRL algorithms, Deep Deterministic Policy Gradient (DDPG) and Soft-Actor Critic (SAC), which can handle the continuous action space of acceleration/deceleration. The proposed models have been trained on the combination of real-world leader vehicle (LV) trajectories and simulated trajectories generated using the Ornstein-Uhlenbeck (OU) process. The overall performance of the proposed models has been tested using Cumulative Distribution Function (CDF) plots and compared with the real-world trajectory data. The results show that the RL models successfully maintain lower distance headway (i.e., higher efficiency) and jerk compared to human-driven vehicles without compromising safety. Further, to assess the robustness of the proposed models, we evaluated the model performance on diverse safety-critical scenarios, in terms of car-following and traffic signal compliance. Both DDPG and SAC models successfully handled the critical scenarios, while the DDPG model showed smoother action profiles compared to the SAC model. Overall, the results confirm that DRL-based longitudinal vehicle control strategy at SI can help to improve traffic safety, efficiency, and comfort.

cs.AI↗

arXiv2025

RiM: Record, Improve and Maintain Physical Well-being using Federated Learning

In academic settings, the demanding environment often forces students to prioritize academic performance over their physical well-being. Moreover, privacy concerns and the inherent risk of data breaches hinder the deployment of traditional machine learning techniques for addressing these health challenges. In this study, we introduce RiM: Record, Improve, and Maintain, a mobile application which incorporates a novel personalized machine learning framework that leverages federated learning to enhance students' physical well-being by analyzing their lifestyle habits. Our approach involves pre-training a multilayer perceptron (MLP) model on a large-scale simulated dataset to generate personalized recommendations. Subsequently, we employ federated learning to fine-tune the model using data from IISER Bhopal students, thereby ensuring its applicability in real-world scenarios. The federated learning approach guarantees differential privacy by exclusively sharing model weights rather than raw data. Experimental results show that the FedAvg-based RiM model achieves an average accuracy of 60.71% and a mean absolute error of 0.91--outperforming the FedPer variant (average accuracy 46.34%, MAE 1.19)--thereby demonstrating its efficacy in predicting lifestyle deficits under privacy-preserving constraints.

cs.LG↗

arXiv2025

SPIRIT: Short-term Prediction of solar IRradIance for zero-shot Transfer learning using Foundation Models

Traditional solar forecasting models are based on several years of site-specific historical irradiance data, often spanning five or more years, which are unavailable for newer photovoltaic farms. As renewable energy is highly intermittent, building accurate solar irradiance forecasting systems is essential for efficient grid management and enabling the ongoing proliferation of solar energy, which is crucial to achieve the United Nations' net zero goals. In this work, we propose SPIRIT, a novel approach leveraging foundation models for solar irradiance forecasting, making it applicable to newer solar installations. Our approach outperforms state-of-the-art models in zero-shot transfer learning by about 70%, enabling effective performance at new locations without relying on any historical data. Further improvements in performance are achieved through fine-tuning, as more location-specific data becomes available. These findings are supported by statistical significance, further validating our approach. SPIRIT represents a pivotal step towards rapid, scalable, and adaptable solar forecasting solutions, advancing the integration of renewable energy into global power systems.

cs.LG↗

arXiv2025

Subject Representation Learning from EEG using Graph Convolutional Variational Autoencoders

We propose GC-VASE, a graph convolutional-based variational autoencoder that leverages contrastive learning for subject representation learning from EEG data. Our method successfully learns robust subject-specific latent representations using the split-latent space architecture tailored for subject identification. To enhance the model's adaptability to unseen subjects without extensive retraining, we introduce an attention-based adapter network for fine-tuning, which reduces the computational cost of adapting the model to new subjects. Our method significantly outperforms other deep learning approaches, achieving state-of-the-art results with a subject balanced accuracy of 89.81% on the ERP-Core dataset and 70.85% on the SleepEDFx-20 dataset. After subject adaptive fine-tuning using adapters and attention layers, GC-VASE further improves the subject balanced accuracy to 90.31% on ERP-Core. Additionally, we perform a detailed ablation study to highlight the impact of the key components of our method.

eess.SP↗

arXiv2025

Patent Novelty Assessment Accelerating Innovation and Patent Prosecution

In the rapidly evolving landscape of technological innovation, safeguarding intellectual property rights through patents is crucial for fostering progress and stimulating research and development investments. This report introduces a ground-breaking Patent Novelty Assessment and Claim Generation System, meticulously crafted to dissect the inventive aspects of intellectual property and simplify access to extensive patent claim data. Addressing a crucial gap in academic institutions, our system provides college students and researchers with an intuitive platform to navigate and grasp the intricacies of patent claims, particularly tailored for the nuances of Chinese patents. Unlike conventional analysis systems, our initiative harnesses a proprietary Chinese API to ensure unparalleled precision and relevance. The primary challenge lies in the complexity of accessing and comprehending diverse patent claims, inhibiting effective innovation upon existing ideas. Our solution aims to overcome these barriers by offering a bespoke approach that seamlessly retrieves comprehensive claim information, finely tuned to the specifics of the Chinese patent landscape. By equipping users with efficient access to comprehensive patent claim information, our transformative platform seeks to ignite informed exploration and innovation in the ever-evolving domain of intellectual property. Its envisioned impact transcends individual colleges, nurturing an environment conducive to research and development while deepening the understanding of patented concepts within the academic community.

cs.DL↗

arXiv2024

Towards Infusing Auxiliary Knowledge for Distracted Driver Detection

Distracted driving is a leading cause of road accidents globally. Identification of distracted driving involves reliably detecting and classifying various forms of driver distraction (e.g., texting, eating, or using in-car devices) from in-vehicle camera feeds to enhance road safety. This task is challenging due to the need for robust models that can generalize to a diverse set of driver behaviors without requiring extensive annotated datasets. In this paper, we propose KiD3, a novel method for distracted driver detection (DDD) by infusing auxiliary knowledge about semantic relations between entities in a scene and the structural configuration of the driver's pose. Specifically, we construct a unified framework that integrates the scene graphs, and driver pose information with the visual cues in video frames to create a holistic representation of the driver's actions.Our results indicate that KiD3 achieves a 13.64% accuracy improvement over the vision-only baseline by incorporating such auxiliary knowledge with visual information.

cs.CV↗

arXiv2024

DECODE: Data-driven Energy Consumption Prediction leveraging Historical Data and Environmental Factors in Buildings

Energy prediction in buildings plays a crucial role in effective energy management. Precise predictions are essential for achieving optimal energy consumption and distribution within the grid. This paper introduces a Long Short-Term Memory (LSTM) model designed to forecast building energy consumption using historical energy data, occupancy patterns, and weather conditions. The LSTM model provides accurate short, medium, and long-term energy predictions for residential and commercial buildings compared to existing prediction models. We compare our LSTM model with established prediction methods, including linear regression, decision trees, and random forest. Encouragingly, the proposed LSTM model emerges as the superior performer across all metrics. It demonstrates exceptional prediction accuracy, boasting the highest R2 score of 0.97 and the most favorable mean absolute error (MAE) of 0.007. An additional advantage of our developed model is its capacity to achieve efficient energy consumption forecasts even when trained on a limited dataset. We address concerns about overfitting (variance) and underfitting (bias) through rigorous training and evaluation on real-world data. In summary, our research contributes to energy prediction by offering a robust LSTM model that outperforms alternative methods and operates with remarkable efficiency, generalizability, and reliability.

cs.LG↗

arXiv2022

Ocean Mover's Distance: Using Optimal Transport for Analyzing Oceanographic Data

Remote sensing observations from satellites and global biogeochemical models have combined to revolutionize the study of ocean biogeochemical cycling, but comparing the two data streams to each other and across time remains challenging due to the strong spatial-temporal structuring of the ocean. Here, we show that the Wasserstein distance provides a powerful metric for harnessing these structured datasets for better marine ecosystem and climate predictions. Wasserstein distance complements commonly used point-wise difference methods such as the root mean squared error, by quantifying differences in terms of spatial displacement in addition to magnitude. As a test case we consider Chlorophyll (a key indicator of phytoplankton biomass) in the North-East Pacific Ocean, obtained from model simulations, in situ measurements, and satellite observations. We focus on two main applications: 1) Comparing model predictions with satellite observations, and 2) temporal evolution of Chlorophyll both seasonally and over longer time frames. Wasserstein distance successfully isolates temporal and depth variability and quantifies shifts in biogeochemical province boundaries. It also exposes relevant temporal trends in satellite Chlorophyll consistent with climate change predictions. Our study shows that optimal transport vectors underlying Wasserstein distance provide a novel visualization tool for testing models and better understanding temporal dynamics in the ocean.

stat.AP↗

arXiv2022

Statistical Mechanical Interpretation of 2D Pauli Crystals

A manifestation of the Pauli Exclusion Principle is observed when fermions are trapped in the ground state of a 2D harmonic oscillator trap at very low temperatures. This non-interaction of fermions results in the formation of Pauli crystals. This work introduces a statistical mechanical interpretation of the principles that led to the observation of Pauli crystals by calculating the energy, radius and other parameters of the non-interacting arrangement of ultra-cold fermions. This model approaches the problem from two different directions, namely fermion degeneracy and harmonic oscillator treatment. Unifying the two different approaches gives a more comprehensive and robust description of the various parameters of Pauli crystals.

cond-mat.stat-mech↗

arXiv2020

Generalized Co-sparse Factor Regression

Multivariate regression techniques are commonly applied to explore the associations between large numbers of outcomes and predictors. In real-world applications, the outcomes are often of mixed types, including continuous measurements, binary indicators, and counts, and the observations may also be incomplete. Building upon the recent advances in mixed-outcome modeling and sparse matrix factorization, generalized co-sparse factor regression (GOFAR) is proposed, which utilizes the flexible vector generalized linear model framework and encodes the outcome dependency through a sparse singular value decomposition (SSVD) of the integrated natural parameter matrix. To avoid the estimation of the notoriously difficult joint SSVD, GOFAR proposes both sequential and parallel unit-rank estimation procedures. By combining the ideas of alternating convex search and majorization-minimization, an efficient algorithm with guaranteed convergence is developed to solve the sparse unit-rank problem and implemented in the R package gofar. Extensive simulation studies and two real-world applications demonstrate the effectiveness of the proposed approach.

stat.ME↗

arXiv2020

Bright perovskite light-emitting electrochemical cell utilizing CNT sheets as a tunable charge injector

Organic-inorganic perovskite light-emitting devices have recently emerged as a reliable light source. Here, we developed a Single Layer Perovskite Light-Emitting Electrochemical Cells (SL-PeLEC) with laminated free-standing Carbon Nanotube Sheet (CNT) sheets as an effective charge electron injecting cathode electrode. The structure consists of bottom ITO-on-glass as a transparent electrode, the composite of CsPbBr3:PEO:LiPF6 with additive ionic salt as an emitting layer (EML) and 5 layers of CNT aerogel sheets as a top laminated cathode . Utilizing CNT free standing sheets laminated right on top of perovskite thin film in this simple single layer configuration has multiple benefits. Such CNT top cathode does not show any chemical degradation by reaction with halogens from perovskite, which is detrimental for metallic cathodes. Moreover, the formation of an internal p-i-n junction in perovskite EML composite layer by ionic migration under applied voltage bias and electric double layer (EDL) formation at each electrode interface is beneficially effecting CNT sheets by Li+ ionic doping and raises their Fermi level, further enhancing electron injection. Besides, inspired by successes of ionic additives in LECs and electrochemical doping of perovskite with alkali metals, we leveraged a lithium salt, LiPF6, within a CsPbBr3:PEO composite matrix to achieve optimal ionic redistribution and doping effects in this SL-PeLEC. Although initially CNT electrode has slightly high sheet resistance, the SL-PeLEC device has a low turn-on voltage of 2.6v and a maximum luminance intensity of 530 cd/m2, confirming the n-doping increased conductivity. This work provides a unique route toward flexible and bright perovskite LECs with stable and transparent CNT electrodes that can have injection efficiency tuned by poling induced ionic EDL-doping

physics.app-ph↗

arXiv2020

Robust Regression with Compositional Covariates

Many biological high-throughput data sets, such as targeted amplicon-based and metagenomic sequencing data, are compositional in nature. A common exploratory data analysis task is to infer statistical associations between the high-dimensional microbial compositions and habitat- or host-related covariates. We propose a general robust statistical regression framework, RobRegCC (Robust Regression with Compositional Covariates), which extends the linear log-contrast model by a mean shift formulation for capturing outliers. RobRegCC includes sparsity-promoting convex and non-convex penalties for parsimonious model estimation, a data-driven robust initialization procedure, and a novel robust cross-validation model selection scheme. We show RobRegCC's ability to perform simultaneous sparse log-contrast regression and outlier detection over a wide range of simulation settings and provide theoretical non-asymptotic guarantees for the underlying estimators. To demonstrate the seamless applicability of the workflow on real data, we consider a gut microbiome data set from HIV patients and infer robust associations between a sparse set of microbial species and host immune response from soluble CD14 measurements. All experiments are fully reproducible and available on GitHub at https://github.com/amishra-stats/robregcc.

stat.ME↗

arXiv2020

Bright and Color-Tunable Single Layer Perovskite Host-Ionic Guest Light Emitting Electrochemical Cells

Perovskite light emitting devices have drawn considerable attention for their favorable optoelectronic properties. High carrier mobilities make perovskites excellent candidates as host materials in electroluminescent devices. To achieve high performance in a simple single layer device, we employed a CsPbBr3 perovskite host and a novel ionic iridium complex guest along with a polyelectrolyte to demonstrate efficient light emitting electrochemical cells (PeLECs). For an optimal guest/host blend, 10600 cd/m2 luminance at 11.6 cd/A and 9.04 Lm/W are achieved at 4.1 V. These devices showed voltage-dependent electroluminescence color proceeding from orange-red to green, facilitated by the reconfigurable ionic materials blend. Optimized devices exhibited stable operation under constant current driving, maintaining >630 cd/m2 emission for 40 h. Our rationally-designed ionic guest at an optimal concentration of the host produced efficient (>90%) Forster energy transfer and improved thin film morphological for high performance PeLEC operation enabled by ionic migration to interfaces

cond-mat.mtrl-sci↗

arXiv2020

Using LSTM for the Prediction of Disruption in ADITYA Tokamak

Major disruptions in tokamak pose a serious threat to the vessel and its surrounding pieces of equipment. The ability of the systems to detect any behavior that can lead to disruption can help in alerting the system beforehand and prevent its harmful effects. Many machine learning techniques have already been in use at large tokamaks like JET and ASDEX, but are not suitable for ADITYA, which is comparatively small. Through this work, we discuss a new real-time approach to predict the time of disruption in ADITYA tokamak and validate the results on an experimental dataset. The system uses selected diagnostics from the tokamak and after some pre-processing steps, sends them to a time-sequence Long Short-Term Memory (LSTM) network. The model can make the predictions 12 ms in advance at less computation cost that is quick enough to be deployed in real-time applications.

cs.LG↗