most citedEPOpt: Learning Robust Neural Network Policies Using Model Ensembles

144 citations · 244 across the 2 of their papers we have counts for

collaborators
Showing cs.LGShow all

2 papers · 1 filter

cs.LG2019★ 100 cited

Online Meta-Learning

Chelsea Finn, Aravind Rajeswaran, Sham Kakade +1

A central capability of intelligent systems is the ability to continuously build upon previous experiences to speed up and enhance learning of new tasks. Two distinct research para…

cs.LG2016★ 144 cited

EPOpt: Learning Robust Neural Network Policies Using Model Ensembles

Aravind Rajeswaran, Sarvjeet Ghotra, Balaraman Ravindran +1

Sample complexity and safety are major challenges when learning policies with reinforcement learning for real-world tasks, especially when the policies are represented using rich f…