Search-Based Testing of Reinforcement Learning
This work addresses the problem of opaque and stochastic RL evaluation for researchers and practitioners, offering incremental improvements in testing methodologies.
The paper tackles the challenge of evaluating deep reinforcement learning agents by introducing a search-based testing framework that identifies safety-critical boundary states and uses fuzz traces for robust performance assessment, applied to Super Mario Bros. with results showing improved safety and performance metrics.
Evaluation of deep reinforcement learning (RL) is inherently challenging. Especially the opaqueness of learned policies and the stochastic nature of both agents and environments make testing the behavior of deep RL agents difficult. We present a search-based testing framework that enables a wide range of novel analysis capabilities for evaluating the safety and performance of deep RL agents. For safety testing, our framework utilizes a search algorithm that searches for a reference trace that solves the RL task. The backtracking states of the search, called boundary states, pose safety-critical situations. We create safety test-suites that evaluate how well the RL agent escapes safety-critical situations near these boundary states. For robust performance testing, we create a diverse set of traces via fuzz testing. These fuzz traces are used to bring the agent into a wide variety of potentially unknown states from which the average performance of the agent is compared to the average performance of the fuzz traces. We apply our search-based testing approach on RL for Nintendo's Super Mario Bros.