Hindsight Foresight Relabeling For Meta-reinforcement Learning

Abstract

Meta-reinforcement learning (meta-RL) algorithms allow for agents to learn new behaviors from small amounts of experience, mitigating the sample inefficiency problem in RL. However, while meta-RL agents can adapt quickly to new tasks at test time after experiencing only a few trajectories, the meta-training process is still sample-inefficient. Prior works have found that in the multi-task RL setting, relabeling past transitions and thus sharing experience among tasks can improve sample efficiency and asymptotic performance. We apply this idea to the meta-RL setting and devise a new relabeling method called Hindsight Foresight Relabeling (HFR). We construct a relabeling distribution using the combination of "hindsight", which is used to relabel trajectories using reward functions from the training task distribution, and "foresight", which takes the relabeled trajectories and computes the utility of each trajectory for each task. HFR is easy to implement and readily compatible with exist

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keywan2021hindsight

Related papers

• Rewriting History With Inverse RL: Hindsight Inference For Policy Improvement (2020)0.00
• Efficient Meta Reinforcement Learning For Preference-based Fast Adaptation (2022)0.00
• Distributionally Adaptive Meta Reinforcement Learning (2022)2.26
• A Tutorial On Meta-reinforcement Learning (2023)10.85
• RL\(^3\): Boosting Meta Reinforcement Learning Via RL Inside RL\(^2\) (2023)0.00
• HMRL: Hyper-meta Learning For Sparse Reward Reinforcement Learning Problem (2020)0.00
• Guided Meta-policy Search (2019)0.00
• Hypothesis Network Planned Exploration For Rapid Meta-reinforcement Learning Adaptation (2023)0.00