SLM Lab: A Comprehensive Benchmark and Modular Software Framework for Reproducible Deep Reinforcement Learning
This work addresses reproducibility issues for researchers in reinforcement learning, though it is incremental as it builds on existing methods and frameworks.
The authors tackled the problem of reproducibility and benchmarking in deep reinforcement learning by introducing SLM Lab, a modular software framework that enables reproducible experiments and includes a comprehensive benchmark of RL algorithms, resulting in the evaluation of a discrete-action variant of Soft Actor-Critic and a hybrid training method.
We introduce SLM Lab, a software framework for reproducible reinforcement learning (RL) research. SLM Lab implements a number of popular RL algorithms, provides synchronous and asynchronous parallel experiment execution, hyperparameter search, and result analysis. RL algorithms in SLM Lab are implemented in a modular way such that differences in algorithm performance can be confidently ascribed to differences between algorithms, not between implementations. In this work we present the design choices behind SLM Lab and use it to produce a comprehensive single-codebase RL algorithm benchmark. In addition, as a consequence of SLM Lab's modular design, we introduce and evaluate a discrete-action variant of the Soft Actor-Critic algorithm (Haarnoja et al., 2018) and a hybrid synchronous/asynchronous training method for RL agents.