Neural network pruning is a technique of network compression by removing weights of lower importance from an optimized neural network. Often, pruned networks are compared in terms of accuracy, which is realized in terms of rewards for Deep Reinforcement Learning (DRL) networks. However, networks that estimate control actions for safety-critical tasks, must also adhere to safety requirements along with obtaining rewards. We propose a methodology to iteratively refine the weights of a pruned neural network such that we get a sparse high-performance network without significant side effects on safety.

Recent advances in Reinforcement Learning (RL) combined with Deep Learning (DL) have demonstrated impressive performance in complex tasks, including autonomous driving. The use of RL agents in autonomous driving leads to a smooth human-like driving experience, but the limited interpretability of Deep Reinforcement Learning (DRL) creates a verification and certification bottleneck. Instead of relying on RL agents to learn complex tasks, we propose HPRL - Hierarchical Program-triggered Reinforcement Learning, which uses a hierarchy consisting of a structured program along with multiple RL agents, each trained to perform a relatively simple task. The focus of verification shifts to the master program under simple guarantees from the RL agents, leading to a significantly more interpretable and verifiable implementation as compared to a complex RL agent. The evaluation of the framework is demonstrated on different driving tasks, and National Highway Traffic Safety Administration (NHTSA) pre-crash scenarios using CARLA, an open-source dynamic urban simulation environment.

Constructing Reinforcement Learning (RL) policies that adhere to safety requirements is an emerging field of study. RL agents learn via trial and error with an objective to optimize a reward signal. Often policies that are designed to accumulate rewards do not satisfy safety specifications. We present a methodology for counterexample guided refinement of a trained RL policy against a given safety specification. Our approach has two main components. The first component is an approach to discover failure trajectories using Bayesian optimization over multiple parameters of uncertainty from a policy learnt in a model-free setting. The second component selectively modifies the failure points of the policy using gradient-based updates. The approach has been tested on several RL environments, and we demonstrate that the policy can be made to respect the safety specifications through such targeted changes.