Hierarchical Programmatic Reinforcement Learning via Learning to Compose Programs
This work addresses the problem of producing interpretable and generalizable reinforcement learning policies for researchers and practitioners, though it is incremental as it builds on existing methods like LEAPS.
The paper tackles the limitations of prior programmatic reinforcement learning methods by proposing a hierarchical framework that learns to compose programs, enabling the generation of out-of-distributionally complex behaviors and more precise credit assignment, with experimental results in the Karel domain showing it outperforms baselines.
Aiming to produce reinforcement learning (RL) policies that are human-interpretable and can generalize better to novel scenarios, Trivedi et al. (2021) present a method (LEAPS) that first learns a program embedding space to continuously parameterize diverse programs from a pre-generated program dataset, and then searches for a task-solving program in the learned program embedding space when given a task. Despite the encouraging results, the program policies that LEAPS can produce are limited by the distribution of the program dataset. Furthermore, during searching, LEAPS evaluates each candidate program solely based on its return, failing to precisely reward correct parts of programs and penalize incorrect parts. To address these issues, we propose to learn a meta-policy that composes a series of programs sampled from the learned program embedding space. By learning to compose programs, our proposed hierarchical programmatic reinforcement learning (HPRL) framework can produce program policies that describe out-of-distributionally complex behaviors and directly assign credits to programs that induce desired behaviors. The experimental results in the Karel domain show that our proposed framework outperforms baselines. The ablation studies confirm the limitations of LEAPS and justify our design choices.