Reinforcement Learning With Trajectory Feedback

Abstract

The standard feedback model of reinforcement learning requires revealing the reward of every visited state-action pair. However, in practice, it is often the case that such frequent feedback is not available. In this work, we take a first step towards relaxing this assumption and require a weaker form of feedback, which we refer to as *trajectory feedback*. Instead of observing the reward obtained after every action, we assume we only receive a score that represents the quality of the whole trajectory observed by the agent, namely, the sum of all rewards obtained over this trajectory. We extend reinforcement learning algorithms to this setting, based on least-squares estimation of the unknown reward, for both the known and unknown transition model cases, and study the performance of these algorithms by analyzing their regret. For cases where the transition model is unknown, we offer a hybrid optimistic-Thompson Sampling approach that results in a tractable algorithm.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyefroni2020reinforcement

Related papers

• Explicitly Encouraging Low Fractional Dimensional Trajectories Via Reinforcement Learning (2020)0.00
• Dueling RL: Reinforcement Learning With Trajectory Preferences (2021)0.00
• On The Theory Of Reinforcement Learning With Once-per-episode Feedback (2021)0.00
• Provably Efficient Offline Reinforcement Learning With Trajectory-wise Reward (2022)0.00
• Harnessing Mixed Offline Reinforcement Learning Datasets Via Trajectory Weighting (2023)0.00
• Learning Long-term Reward Redistribution Via Randomized Return Decomposition (2021)0.00
• Memory Based Trajectory-conditioned Policies For Learning From Sparse Rewards (2019)0.00
• Inverse Reinforcement Learning From Non-stationary Learning Agents (2024)0.00