Berkeley AI Research Lab University of California, Berkeley
My research goal is to develop agents that can learn via interaction, reliably, and robustly in the real world. Towards this goal, I am interested in devising effective algorithms for sequential decision-making by effectively leveraging previously collected static datasets (offline, data-driven RL) that can serve as rich sources of experience, understanding and addressing challenges in applying RL with deep neural network function approximation, and making it easy to actually apply RL in practice, by tackling problems such as hyperparameter tuning. I am also particularly interested in applying these algorithms to actual decision-making problem, and I am actively working towards many applications – if you are interestd, please check out our recent work on offline decision-making for chip design! I am a receipient of the Apple Scholars in AI/ML PhD Fellowship, 2022.
论文共 48 篇作者统计合作学者相似作者
, , ,,,,, ,