Programmatically Interpretable Reinforcement Learning

Abhinav Verma; Vijayaraghavan Murali; Rishabh Singh; Pushmeet Kohli, Swarat Chaudhuri

We present a reinforcement learning framework, called Programmatically Interpretable Reinforcement Learning (PIRL), that is designed to generate interpretable and verifiable agent policies. Unlike the popular Deep Reinforcement Learning (DRL) paradigm, which represents policies by neural networks, PIRL represents policies using a high-level, domain-specific programming language. Such programmatic policies have the benefits of being more easily interpreted than neural networks, and being

more »

... to verification by symbolic methods. We propose a new method, called Neurally Directed Program Search (NDPS), for solving the challenging nonsmooth optimization problem of finding a programmatic policy with maximal reward. NDPS works by first learning a neural policy network using DRL, and then performing a local search over programmatic policies that seeks to minimize a distance from this neural "oracle". We evaluate NDPS on the task of learning to drive a simulated car in the TORCS car-racing environment. We demonstrate that NDPS is able to discover human-readable policies that pass some significant performance bars. We also show that PIRL policies can have smoother trajectories, and can be more easily transferred to environments not encountered during training, than corresponding policies discovered by DRL.

<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1804.02477v3">arXiv:1804.02477v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3hydz34zurb4xcr35v6hlkmuvi">fatcat:3hydz34zurb4xcr35v6hlkmuvi</a> </span>

<span class="ui label basic blue small">Multiple Versions</span>

<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200914203745/https://arxiv.org/pdf/1804.02477v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/36/9d/369d450fa9f137c271141a8da7c7813a71f27e00.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1804.02477v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Programmatically Interpretable Reinforcement Learning [article]

Preserved Fulltext

Other Versions