Publications

We present a formal operator-theoretic framework for analyzing Transformer-based language models using free probability theory. By modeling token embeddings and attention mechanisms as self-adjoint

Clustering stands as one of the most prominent challenges in unsupervised machine learning. Among centroid-based methods, the classic k-means algorithm, based on Lloyd’s heuristic, is

The Gromov-Wasserstein (GW) distance serves as a measure of discrepancy between two distributions that are supported on distinct ambient spaces. Emerging as the optimal expected

Uncertainty characterization via posteriors followed by Bayesian updates is an acclaimed way to aid the exploration of model-free Reinforcement Learning (RL) algorithms. Motivated by the

A long-tail class imbalanced learning problem is a scenario where the rare or minority classes, representing infrequent events or categories, make up the long tail

Amongst the numerous variants Variational Autoencoder (VAE) has inspired, the Wasserstein Autoencoder (WAE) stands out due to its heightened generative quality and intriguing theoretical properties.

Imbalanced image datasets pose significant challenges for developing robust classifiers, particularly when certain classes are heavily underrepresented. To tackle this issue, we propose Density-Driven Attraction

Graph neural networks (GNNs) witness impressive performances on homophilic graphs characterized by a higher number of edges connecting nodes of similar class labels. A decline

This study addresses the challenges of learning from long-tailed class imbalances in deep neural networks, particularly for image recognition. Long-tailed class imbalances occur when a

Data can be represented by multiple sets of features, where each semantically coherent set of features is called a view. For example, an image can

Accurate breast tumor segmentation and malignancy detection are crucial for early cancer diagnosis. In this context, we propose a novel lightweight multi-task learning framework, MA-DTNet,

The early detection of neurodegenerative disorders such as Alzheimer’s disease is crucial to providing effective healthcare for management and recovery. We address the task of

Contrastive Learning (CL) aims to create effective embeddings for input data by minimizing the distance between positive pairs, i.e., different augmentations or views of the

Medical image segmentation serves as a critical tool for healthcare professionals, enabling the precise extraction of Regions of Interest (ROIs) from clinical images at the

Deep learning models have shown considerable promise in the classification of skin lesions. However, a notable challenge arises from their inherent bias towards dominant skin

Supervised classification problems from the real world typically face a challenge characterized by the scarcity of samples in one or more target classes compared to

Class imbalance in the medical image dataset is almost inherent due to the limited availability of clinical data for certain diseases and patient populations. Under-represented

The homophily (heterophily) ratio in a graph represents the proportion of edges connecting nodes with similar (dissimilar) class labels. Existing methods for estimating the homophily

Since its inception, Generative Adversarial Networks (GAN) have marked a triumph in generative modeling. Its impeccable capacity to mimic observations from unknown probability distributions has