Solving Compositional Reinforcement Learning Problems via Task Reduction

Abstract

We propose a novel learning paradigm, Self-Imitation via Reduction (SIR), for solving compositional reinforcement learning problems. SIR is based on two core ideas: task reduction and self-imitation. Task reduction tackles a hard-to-solve task by actively reducing it to an easier task whose solution is known by the RL agent. Once the original hard task is successfully solved by task reduction, the agent naturally obtains a self-generated solution trajectory to imitate. By continuously collecting and imitating such demonstrations, the agent is able to progressively expand the solved subspace in the entire task space. Experiment results show that SIR can significantly accelerate and improve learning on a variety of challenging sparse-reward continuous-control problems with compositional structures. Code and videos are available at https://sites.google.com/view/sir-compositional.

Project website link updated

References in corpus (8)

• Practical Bayesian Optimization of Machine Learning Algorithms
• Language Models are Few-Shot Learners
• FeUdal Networks for Hierarchical Reinforcement Learning
• Freeze-Thaw Bayesian Optimization
• Planning with Goal-Conditioned Policies
• Compositional Planning Using Optimal Option Models
• Learning to Compose Skills
• Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments