Jayaraman J. Thiagarajan

37 accepted papers

2025

Leveraging Registers in Vision Transformers for Robust Adaptation

Srikar Yellapragada, Kowshik Thopalli, Vivek Sivaraman Narayanaswamy, Wesam A. Sakla, Yang Liu, Yamen Mubarka +2

ICASSP 2025accepted

Vision Transformers (ViTs) have shown success across a variety of tasks due to their ability to capture global image representations. Recent studies have identified the existence of high-norm tokens in ViTs, which can interfere with unsupervised object discovery. To address this, the use of "registe…

Cited by 3 Source ↗Scholar ↗

2025

On The Role of Prompt Construction In Enhancing Efficacy and Efficiency of LLM-Based Tabular Data Generation

Banooqa H. Banday, Kowshik Thopalli, Tanzima Z. Islam, Jayaraman J. Thiagarajan

ICASSP 2025accepted

LLM-based data generation for real-world tabular data can be challenged by the lack of sufficient semantic context in feature names used to describe columns. We hypothesize that enriching prompts with even minimal contextual information, such as a brief explanation of what each feature represents ca…

Cited by 0 Source ↗Scholar ↗

2024

Accurate and Scalable Estimation of Epistemic Uncertainty for Graph Neural Networks

Puja Trivedi, Mark Heimann, Rushil Anirudh, Danai Koutra, Jayaraman J. Thiagarajan

ICLR 2024poster

While graph neural networks (GNNs) are widely used for node and graph representation learning tasks, the reliability of GNN uncertainty estimates under distribution shifts remains relatively under-explored. Indeed, while post-hoc calibration strategies can be used to improve in-distribution calibrat…

Cited by 4 Source ↗PDF ↗Scholar ↗

2024

DECIDER: Leveraging Foundation Model Priors for Improved Model Failure Detection and Explanation

Rakshith Subramanyam*, Kowshik Thopalli*, Vivek Sivaraman Narayanaswamy, Jayaraman J. Thiagarajan

ECCV 2024poster

"Reliably detecting when a deployed machine learning model is likely to fail on a given input is crucial for ensuring safe operation. In this work, we propose DECIDER (Debiasing Classifiers to Identify Errors Reliably), a novel approach that leverages priors from large language models (LLMs) and vis…

Cited by 3 Source ↗PDF ↗code ↗Scholar ↗

2024

Exploring the Utility of Clip Priors for Visual Relationship Prediction

Rakshith Subramanyam, T. S. Jayram, Rushil Anirudh, Jayaraman J. Thiagarajan

ICASSP 2024accepted

This work explores the challenges of leveraging large-scale vision language models, such as CLIP, for visual relationship prediction (VRP), a task vital in understanding the relations between objects in a scene based on both image features and text descriptors. Despite its potential, we find that CL…

Cited by 0 Source ↗Scholar ↗

2024

On Estimating Link Prediction Uncertainty Using Stochastic Centering

Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

ICASSP 2024accepted

Accurate confidence estimates are crucial for safe graph neural network (GNN) deployment, yet link prediction (LP) calibration is understudied. We provide novel insights into LP calibration by highlighting the importance of meaningful node-level uncertainties. In response, we propose E-ΔUQ, an archi…

Cited by 0 Source ↗Scholar ↗

2024

On the Use of Anchoring for Training Vision Models

Vivek Narayanaswamy, Kowshik Thopalli, Rushil Anirudh, Yamen Mubarka, Wesam A. Sakla, Jayaraman J. Thiagarajan

NeurIPS 2024spotlight

Anchoring is a recent, architecture-agnostic principle for training deep neural networks that has been shown to significantly improve uncertainty estimation, calibration, and extrapolation capabilities. In this paper, we systematically explore anchoring as a general protocol for training vision mode…

Cited by 0 Source ↗PDF ↗Scholar ↗

2024

PAGER: Accurate Failure Characterization in Deep Regression Models

Jayaraman J. Thiagarajan, Vivek Narayanaswamy, Puja Trivedi, Rushil Anirudh

ICML 2024poster

Safe deployment of AI models requires proactive detection of failures to prevent costly errors. To this end, we study the important problem of detecting failures in deep regression models. Existing approaches rely on epistemic uncertainty estimates or inconsistency w.r.t the training data to identif…

Cited by 3 Source ↗PDF ↗Scholar ↗

2024

The Double-Edged Sword Of Ai Safety: Balancing Anomaly Detection and OOD Generalization Via Model Anchoring

Vivek Sivaraman Narayanaswamy, Rushil Anirudh, Jayaraman J. Thiagarajan

ICASSP 2024accepted

Safe deployment of AI systems requires models to accurately flag anomalous or semantically unrelated data, while also generalizing to unseen shifts in the data distribution. While both these problems have been extensively studied, there is a risk for undesirable trade-off when exclusively optimizing…

Cited by 0 Source ↗Scholar ↗

2023

A Closer Look At Scoring Functions And Generalization Prediction

Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

ICASSP 2023accepted

Generalization error predictors (GEPs) aim to predict model performance on unseen distributions by deriving dataset-level error estimates from sample-level scores. However, GEPs often utilize disparate mechanisms (e.g., regressors, thresholding functions, calibration datasets, etc), to derive such e…

Cited by 0 Source ↗Scholar ↗

2023

A Closer Look at Model Adaptation using Feature Distortion and Simplicity Bias

Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

ICLR 2023top-25%

Advances in the expressivity of pretrained models have increased interest in the design of adaptation protocols which enable safe and effective transfer learning. Going beyond conventional linear probing (LP) and fine tuning (FT) strategies, protocols that can effectively control feature distortion,…

Cited by 24 Source ↗PDF ↗Scholar ↗

2023

Cross-GAN Auditing: Unsupervised Identification of Attribute Level Similarities and Differences Between Pretrained Generative Models

Matthew L. Olson, Shusen Liu, Rushil Anirudh, Jayaraman J. Thiagarajan, Peer-Timo Bremer, Weng-Keen Wong

CVPR 2023poster

Generative Adversarial Networks (GANs) are notoriously difficult to train especially for complex distributions and with limited data. This has driven the need for interpretable tools to audit trained networks, for example, to identify biases or ensure fairness. Existing GAN audit tools are restricte…

Cited by 8 Source ↗PDF ↗code ↗Scholar ↗

2023

DOLCE: A Model-Based Probabilistic Diffusion Framework for Limited-Angle CT Reconstruction

Jiaming Liu, Rushil Anirudh, Jayaraman J. Thiagarajan, Stewart He, K Aditya Mohan, Ulugbek S. Kamilov +1

ICCV 2023poster

Limited-Angle Computed Tomography (LACT) is a non-destructive 3D imaging technique used in a variety of applications ranging from security to medicine. The limited angle coverage in LACT is often a dominant source of severe artifacts in the reconstructed images, making it a challenging imaging inver…

Cited by 88 PDF ↗code ↗Scholar ↗

2023

Single-Shot Domain Adaptation via Target-Aware Generative Augmentations

Rakshith Subramanyam, Kowshik Thopalli, Spring Berman, Pavan K. Turaga, Jayaraman J. Thiagarajan

ICASSP 2023accepted

The problem of adapting models from a source domain using data from any target domain of interest has gained prominence, thanks to the brittle generalization in deep neural networks. While several test-time adaptation techniques have emerged, they typically rely on synthetic data augmentations in ca…

Cited by 0 Source ↗Scholar ↗

2023

Target-Aware Generative Augmentations for Single-Shot Adaptation

Kowshik Thopalli, Rakshith Subramanyam, Pavan K. Turaga, Jayaraman J. Thiagarajan

ICML 2023poster

In this paper, we address the problem of adapting models from a source domain to a target domain, a task that has become increasingly important due to the brittle generalization of deep neural networks. While several test-time adaptation techniques have emerged, they typically rely on synthetic tool…

Cited by 6 Source ↗PDF ↗code ↗Scholar ↗

2022

Analyzing Data-Centric Properties for Graph Contrastive Learning

Puja Trivedi, Ekdeep Singh Lubana, Mark Heimann, Danai Koutra, Jayaraman J. Thiagarajan

NeurIPS 2022accept

Recent analyses of self-supervised learning (SSL) find the following data-centric properties to be critical for learning good representations: invariance to task-irrelevant semantics, separability of classes in some latent space, and recoverability of labels from augmented samples. However, given th…

Cited by 12 Source ↗PDF ↗code ↗Scholar ↗

2022

Improved StyleGAN-v2 based Inversion for Out-of-Distribution Images

Rakshith Subramanyam, Vivek Narayanaswamy, Mark Naufel, Andreas Spanias, Jayaraman J. Thiagarajan

ICML 2022spotlight

Inverting an image onto the latent space of pre-trained generators, e.g., StyleGAN-v2, has emerged as a popular strategy to leverage strong image priors for ill-posed restoration. Several studies have showed that this approach is effective at inverting images similar to the data used for training. H…

Cited by 7 Source ↗PDF ↗code ↗Scholar ↗

2022

Predicting the Generalization Gap in Deep Models using Anchoring

Vivek Sivaraman Narayanaswamy, Rushil Anirudh, Irene Kim, Yamen Mubarka, Andreas Spanias, Jayaraman J. Thiagarajan

ICASSP 2022accepted

We address the problem of predicting the generalization gap of deep neural networks under large, natural, and synthetic distribution shifts between source and target domains. This is crucial in understanding how models behave in uncontrollable ‘in-the-wild’ scenarios, but existing techniques fail wh…

Cited by 0 Source ↗Scholar ↗

2022

Single Model Uncertainty Estimation via Stochastic Data Centering

Jayaraman J. Thiagarajan, Rushil Anirudh, Vivek Narayanaswamy, Peer-timo Bremer

NeurIPS 2022accept

We are interested in estimating the uncertainties of deep neural networks, which play an important role in many scientific and engineering problems. In this paper, we present a striking new finding that an ensemble of neural networks with the same weight initialization, trained on datasets that are…

Cited by 33 Source ↗PDF ↗code ↗Scholar ↗

2022

Sparsity Improves Unsupervised Attribute Discovery in Stylegan

Shusen Liu, Rushil Anirudh, Jayaraman J. Thiagarajan, Peer-Timo Bremer

ICASSP 2022accepted

Rich semantics exist in latent spaces inferred using deep generative models. The ability to extract and interpret them is not only essential for understanding the underlying factors of variation in the data distribution, but also crucial for con-trolled image generation. Several methods have been pr…

Cited by 0 Source ↗Scholar ↗

2021

Accurate and Robust Feature Importance Estimation under Distribution Shifts

Jayaraman J. Thiagarajan, Vivek Narayanaswamy, Rushil Anirudh, Peer-Timo Bremer, Andreas Spanias

AAAI 2021technical

With increasing reliance on the outcomes of black-box models in critical applications, post-hoc explainability tools that do not require access to the model internals are often used to enable humans understand and trust these models. In particular, we focus on the class of methods that can reveal th…

Cited by 18 Source ↗PDF ↗code ↗Scholar ↗

2021

Attribute-Guided Adversarial Training for Robustness to Natural Perturbations

Tejas Gokhale, Rushil Anirudh, Bhavya Kailkhura, Jayaraman J. Thiagarajan, Chitta Baral, Yezhou Yang

AAAI 2021technical

While existing work in robust deep learning has focused on small pixel-level norm-based perturbations, this may not account for perturbations encountered in several real world settings. In many such cases although test data might not be available, broad specifications about the types of perturbation…

Cited by 45 Source ↗PDF ↗code ↗Scholar ↗

2021

Designing Counterfactual Generators using Deep Model Inversion

Jayaraman J. Thiagarajan, Vivek Narayanaswamy, Deepta Rajan, Jia Liang, Akshay Chaudhari, Andreas Spanias

NeurIPS 2021poster

Explanation techniques that synthesize small, interpretable changes to a given image while producing desired changes in the model prediction have become popular for introspecting black-box models. Commonly referred to as counterfactuals, the synthesized explanations are required to contain discernib…

Cited by 27 Source ↗PDF ↗Scholar ↗

2021

Uncertainty-Matching Graph Neural Networks to Defend Against Poisoning Attacks

Uday Shankar Shanthamallu, Jayaraman J. Thiagarajan, Andreas Spanias

AAAI 2021technical

Graph Neural Networks (GNNs), a generalization of neural networks to graph-structured data, are often implemented using message passes between entities of a graph. While GNNs are effective for node classification, link prediction and graph classification, they are vulnerable to adversarial attacks,…

Cited by 19 Source ↗PDF ↗Scholar ↗

2021

Using Deep Image Priors to Generate Counterfactual Explanations

Vivek Sivaraman Narayanaswamy, Jayaraman J. Thiagarajan, Andreas Spanias

ICASSP 2021accepted

Through the use of carefully tailored convolutional neural network architectures, a deep image prior (DIP) can be used to obtain pre-images from latent representation encodings. Though DIP inversion has been known to be superior to conventional regularized inversion strategies such as total variatio…

Cited by 0 Source ↗Scholar ↗

2020

A Regularized Attention Mechanism for Graph Attention Networks

Uday Shankar Shanthamallu, Jayaraman J. Thiagarajan, Andreas Spanias

ICASSP 2020accepted

Machine learning models that can exploit the inherent structure in data have gained prominence. In particular, there is a surge in deep learning solutions for graph-structured data, due to its wide-spread applicability in several fields. Graph attention networks (GAT), a recent addition to the broad…

Cited by 0 Source ↗Scholar ↗

2020

Learn-By-Calibrating: Using Calibration As A Training Objective

Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan

ICASSP 2020accepted

Calibration error is commonly adopted for evaluating the quality of uncertainty estimators in deep neural networks. In this paper, we argue that such a metric is highly beneficial for training predictive models, even when we do not explicitly measure the uncertainties. This is conceptually similar t…

Cited by 0 Source ↗Scholar ↗

2019

Bootstrapping Graph Convolutional Neural Networks for Autism Spectrum Disorder Classification

Rushil Anirudh, Jayaraman J. Thiagarajan

ICASSP 2019accepted

Using predictive models to identify patterns that can act as biomarkers for different neuropathoglogical conditions is becoming highly prevalent. In this paper, we consider the problem of Autism Spectrum Disorder (ASD) classification where previous work has shown that it can be beneficial to incorpo…

Cited by 0 Source ↗Scholar ↗

2019

Designing an Effective Metric Learning Pipeline for Speaker Diarization

Vivek Sivaraman Narayanaswamy, Jayaraman J. Thiagarajan, Huan Song, Andreas Spanias

ICASSP 2019accepted

State-of-the-art speaker diarization systems utilize knowledge from external data, in the form of a pre-trained distance metric, to effectively determine relative speaker identities to unseen data. However, much of recent focus has been on choosing the appropriate feature extractor, ranging from pre…

Cited by 15 Source ↗Scholar ↗

2019

Multiple Subspace Alignment Improves Domain Adaptation

Kowshik Thopalli, Rushil Anirudh, Jayaraman J. Thiagarajan, Pavan K. Turaga

ICASSP 2019accepted

We present a novel unsupervised domain adaptation (DA) method for cross-domain visual recognition. Though subspace methods have found success in DA, their performance is often limited due to the assumption of approximating an entire dataset using a single low-dimensional subspace. Instead, we develo…

Cited by 0 Source ↗Scholar ↗

2019

Understanding Deep Neural Networks through Input Uncertainties

Jayaraman J. Thiagarajan, Irene Kim, Rushil Anirudh, Peer-Timo Bremer

ICASSP 2019accepted

Techniques for understanding the functioning of complex machine learning models are becoming increasingly popular, not only to improve the validation process, but also to extract new insights about the data via exploratory analysis. Though a large class of such tools currently exists, most assume th…

Cited by 0 Source ↗Scholar ↗

2018

Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion

Rushil Anirudh, Hyojin Kim, Jayaraman J. Thiagarajan, K. Aditya Mohan, Kyle Champley, Timo Bremer

CVPR 2018poster

Computed Tomography (CT) reconstruction is a fundamental component to a wide variety of applications ranging from security, to healthcare. The classical techniques require measuring projections, called sinograms, from a full 180 degree view of the object. However, obtaining a full-view is not always…

Cited by 149 Source ↗PDF ↗Scholar ↗

2017

A deep learning approach to multiple kernel fusion

Huan Song, Jayaraman J. Thiagarajan, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy, Andreas Spanias

ICASSP 2017accepted

Kernel fusion is a popular and effective approach for combining multiple features that characterize different aspects of data. Traditional approaches for Multiple Kernel Learning (MKL) attempt to learn the parameters for combining the kernels through sophisticated optimization procedures. In this pa…

Cited by 0 Source ↗Scholar ↗

2016

Beyond L2-loss functions for learning sparse models

Karthikeyan Natesan Ramamurthy, Aleksandr Y. Aravkin, Jayaraman J. Thiagarajan

ICASSP 2016accepted

In sparse learning, the squared Euclidean distance is a popular choice for measuring the approximation quality. However, the use of other forms of parametrized loss functions, including asymmetric losses, has generated research interest. In this paper, we perform sparse learning using a broad class…

Cited by 0 Source ↗Scholar ↗

2016

Consensus inference on mobile phone sensors for activity recognition

Huan Song, Jayaraman J. Thiagarajan, Karthikeyan Natesan Ramamurthy, Andreas Spanias, Pavan K. Turaga

ICASSP 2016accepted

The pervasive use of wearable sensors in activity and health monitoring presents a huge potential for building novel data analysis and prediction frameworks. In particular, approaches that can harness data from a diverse set of low-cost sensors for recognition are needed. Many of the existing approa…

Cited by 0 Source ↗Scholar ↗

2016

Theoretical guarantees for poisson disk sampling using pair correlation function

Bhavya Kailkhura, Jayaraman J. Thiagarajan, Peer-Timo Bremer, Pramod K. Varshney

ICASSP 2016accepted

In this paper, we study the problem of generating uniform random point samples on a domain of d dimensional space based on a minimum distance criterion between point samples (Poisson-disk sampling or PDS). First, we formally define PDS via the pair correlation function (PCF) to quantitatively evalua…

Cited by 0 Source ↗Scholar ↗

2015

Subspace learning using consensus on the grassmannian manifold

Jayaraman J. Thiagarajan, Karthikeyan Natesan Ramamurthy

ICASSP 2015accepted

High-dimensional structure of data can be explored and task-specific representations can be obtained using manifold learning and low-dimensional embedding approaches. However, the uncertainties in data and the sensitivity of the algorithms to parameter settings, reduce the reliability of such repres…

Cited by 0 Source ↗Scholar ↗