About this document

Explainable Reinforcement Learning Review by Carlo Metta is a document available to read on EtoBox.

This document discusses explainability in deep reinforcement learning (XRL). It reviews recent works that aim to make reinforcement learning models more explainable. Explainable techniques can help users understand and trust reinforcement learning models, which are often seen as "black boxes". The document divides explainability techniques for reinforcement learning into two categories: techniques that use transparent algorithms, and post-hoc explainability techniques applied after training. It argues that

Author

Carlo Metta

Language

EN