11 citations · 21 across the 19 of their papers we have counts for
1 paper · 2 filters
Yash Shukla, Abhishek Kulkarni, Robert Wright +2
Despite advances in Reinforcement Learning, many sequential decision making tasks remain prohibitively expensive and impractical to learn. Recently, approaches that automatically g…