发帖
雷达卡
Rewriting History with Inverse RL:
Hindsight Inference for Policy Improvement
Benjamin Eysenbachφθ Xinyang Gengψ Sergey Levineψθ Ruslan Salakhutdinovφ
φ ψ θ
Carnegie Mellon University UC Berkeley Google Brain
Abstract
Multi-task reinforcement learning (RL) aims to simultaneously learn policies for
solving many tasks. Several prior works have found that relabeling past experience
...
京公网安备 11010802022788号
