|
I am a PhD candidate in CSEE at University of Maryland, Baltimore County (UMBC), where I am co advised by Tinoosh Mohsenin and Tim Oates. I am also a visiting researcher at Johns Hopkins University. Before this, I spent time at eBay/PayPal as a software enginner where I worked on PayPal Credit and PayPal Working Capital. I did my MS also at UMBC and BE at University of Pune. Email / Resume / Google Scholar / LinkedIn / GitHub |
|
The goal of my research is to build language guided hierarchical agents capable of solving complex long horizon tasks using RL and human feedback. More recently, I am interested in exploiting large language models to augment these agents.. |
|
Using large pretrained language models (LLMs) to inject common sense priors into hierarchical agents |
|
An interpretable hierarchical agent framework by combining sub-goal generation using language and semantic goal directed reinforcement learning |
|
NeurIPS MineRL BASALT Challenge Winners |
|
A hierarchical agent framework where low-level sub-goals are specified using language |
|
A method for goal generation using dynamical distance functions thus automatically producing a curriculum. |
|
A framework to train RL agents conditioned on constraints that are in the form of structured language |
|
A hybrid architecture for reducing the human intervention time and improving safety by combining model-based and model-free methods. |
|
Learning image representations on unannotated Chest Xray images using the method described in Noroozi and Favaro to gain improvements in classification tasks. Here we they use the pretext task of solving jigsaw puzzles to pre-train the convolutional neural network. Chest x-ray images from the x-ray14 database were used. |
|
Program Committee, Reviewer |
|
|