About Me

I am a PhD student working at the UMass Amherst Computer Vision Lab under the supervision of Prof. Subhransu Maji on data efficient learning algorithms. My previous projects include analyzing generative and contrastive models for discriminative tasks, using coarse supervision for improving segmentation, contrastive learning using part-based clustering. Recently, I have been working on enhancing zero-shot generalization of multimodal models. My interests include improving generative or discriminative models using various types or modes of supervision.

Previously, I was a Research Fellow at Microsoft Research, working with Dr. Prateek Jain and Dr. Harsha Simhadri on the EdgeML project. I worked on developing tiny and accurate Computer Vision models which can be deployed on severely resource-constrained devices. I did my Bachelors and Masters in Tech. in Electronics and Electrical Communication Engineering with a minor in Computer Science and Engineering from Indian Institute of Technology Kharagpur.

Research Interests

Self, Semi and Unsupervised learning, Multimodal Learning, Generative modeling