Jubayer Ibn Hamid

I am a researcher at Stanford Artificial Intelligence Laboratory (SAIL) advised by Chelsea Finn. I am pursuing a B.S in Mathematical Physics and M.S in Computer Science at Stanford University.

My research focuses on the intersection of machine learning, offline reinforcement learning and representation learning. I am also interested in pure mathematics such as abstract algebra, algebraic/differential topology/geometry.

I was born and raised in the beautiful city of Dhaka, Bangladesh. I am a diehard fan of FC Barcelona.

CV / Google Scholar / Email / Twitter

Research

I am currently focused on two main areas: offline policy evaluation, specifically policies trained using behavioral cloning or offline RL, and test-time policy decoding methods, particularly in sampling more optimal strategies in a coherent manner.

(*) denotes co-first authorship

	Bidirectional Decoding: Improving Action Chunking via Closed-Loop Resampling Yuejiang Liu, Jubayer Ibn Hamid, Annie Xie, Yoonho Lee, Max Du, Chelsea Finn (under review). Paper / Website / Code / Blog
	Offline Evaluation of Robotic Manipulation Policies Jubayer Ibn Hamid, Yuejiang Liu, Michał Zawalski, Yoonho Lee, Karl Pertsch, Sergey Levine, Chelsea Finn. (In preparation).
	Tripod: Three Complementary Inductive Biases for Disentangled Representation Learning Kyle Hsu, Jubayer Ibn Hamid, Kaylee Burns, Chelsea Finn, Jiajun Wu ICML, 2024. Paper / Code
	What Makes Pre-trained Visual Representations Successful For Robust Manipulation? Kaylee Burns, Zach Witzel, Jubayer Ibn Hamid, Tianhe Yu, Chelsea Finn, Karol Hausman CoRL, 2024. Paper / Website

Notes

I am sharing notes on various topics that have fascinated me. These are not meant to be in-depth. Rather, they are meant to cover some of the basic constructions that show up periodically and are also interesting in and of themselves.

Algebraic Geometry. Foundational constructions and results in algebraic geometry. (Not finished latex-ing yet, I am updating these on a bi-weekly basis).

Whitney's Embedding Theorem and Immersion Theorem. Weak versions of Whitney's Embedding and Immersion theorem, which are often sufficient.

Policy Gradient Methods. Building blocks (including the policy gradient theorems for both episodic and continuing tasks) of policy gradient algorithms.

Template