Goal-Conditioned Supervised Learning with Sub-Goal Prediction
This work addresses data efficiency in goal-conditioned reinforcement learning for AI agents, presenting an incremental improvement over GCSL.
The paper tackles the limitation of Goal-Conditioned Supervised Learning (GCSL) by proposing Trajectory Iterative Learner (TraIL), which predicts sub-goals from trajectories to enhance data efficiency; results show that TraIL allows agents to reach a greater set of goal states using the same data as GCSL, improving overall performance.
Recently, a simple yet effective algorithm -- goal-conditioned supervised-learning (GCSL) -- was proposed to tackle goal-conditioned reinforcement-learning. GCSL is based on the principle of hindsight learning: by observing states visited in previously executed trajectories and treating them as attained goals, GCSL learns the corresponding actions via supervised learning. However, GCSL only learns a goal-conditioned policy, discarding other information in the process. Our insight is that the same hindsight principle can be used to learn to predict goal-conditioned sub-goals from the same trajectory. Based on this idea, we propose Trajectory Iterative Learner (TraIL), an extension of GCSL that further exploits the information in a trajectory, and uses it for learning to predict both actions and sub-goals. We investigate the settings in which TraIL can make better use of the data, and discover that for several popular problem settings, replacing real goals in GCSL with predicted TraIL sub-goals allows the agent to reach a greater set of goal states using the exact same data as GCSL, thereby improving its overall performance.