Abhranil Chandra
RL PhD, UMass Amherst | Previously @ MILA, GDM, MSR | UWaterloo CS ’25 | IIT KGP ’23 | DeepRL, Self-Improvement, Open-Endedness for Generalist Autonomous Agents
- Role
- Graduate Research Assistant at Manning College Of Information And Computer Sciences, Umass Amherst
- Location
- Amherst, MA, US
- LinkedIn followers
- 500 followers
About Abhranil Chandra
I am a Masters in Computer Science at the University of Waterloo. I recently completed my undergraduate studies at IIT Kharagpur. My research is focused on reinforcement learning, representation learning, graphical and generative models, and theoretical deep learning with applications in NLP and Vision. I am currently working at the intersection of reinforcement learning and representation learning to improve efficiency and OOD generalization of RL algorithms using better learned unsupervised representations.In my undergraduate thesis, I focused on the theoretical analysis of uncertainty estimation and explainability in neural networks using the Bayesian frameworks (Bayesian NNs and Evidential NNs) and developed better-calibrated image classification and segmentation models. I have also worked on building a joint framework for Continual Learning and Domain Generalization using GMM-based internal joint representation consolidation, on multimodal robust representation learning to avoid adversarial visual illusions in depth estimation and other downstream vision tasks, and on large language model based QA systems and natural language generation. I am deeply interested, involved, and invested in the quest for true intelligent agents primarily working towards this goal through the lens of DeepRL, Deep Unsupervised Learning, Causality, and Probabilistic Graphical Models.If you have any queries or opportunities for collaboration, please feel free to reach out to me at my email id- a••••••••@gmail.com.
Experience
Graduate Research Assistant
Manning College Of Information And Computer Sciences, Umass Amherst
Sep 2025 — Present
Advisor: Prof. Scott NiekumFundamentals of Scaling Data-Driven Sequential Decision Making and DeepRL [Unsupervised RL, Offline RL, RL and Interactive Learning beyond Rewards], RL for Foundation Model Agents, Foundation Models for Decision Making, Reasoning, and System-2 Tasks.
Education
University of Waterloo
Research Master's degree , Computer Science (Thesis)
University of Massachusetts Amherst
Doctor of Philosophy - PhD, Computer Science
Indian Institute of Technology, Kharagpur
Bachelor of Technology- BTech (Honors), Mechanical Engineering, with Minor in MnC and Micro-Spl in AI
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.