Explainable Deep Reinforcement Learning

Explainable Deep Reinforcement Learning (XDRL) aims to make the decision-making processes of deep reinforcement learning agents more transparent and understandable. Current research focuses on developing methods to extract interpretable representations from complex neural networks, often employing techniques like policy distillation, feature importance analysis (e.g., SHAP, LIME), and part-based representations, alongside algorithms such as Proximal Policy Optimization (PPO) and Deep Deterministic Policy Gradient (DDPG). This field is crucial for building trust and ensuring the safe and reliable deployment of AI agents in high-stakes applications like finance, healthcare (e.g., drug dosing), and autonomous systems (e.g., UAV control), where understanding the reasoning behind AI decisions is paramount.

Papers

August 21, 2024

Using Part-based Representations for Explainable Deep Reinforcement Learning
Manos Kirtas, Konstantinos Tsampazis, Loukia Avramelou, Nikolaos Passalis, Anastasios Tefas
Deep Reinforcement Learning Part Aware Part Based Negative Training Explainable Deep Reinforcement Learning

July 19, 2024

Explainable Post hoc Portfolio Management Financial Policy of a Deep Reinforcement Learning agent
Alejandra de la Rica Escudero, Eduardo C. Garrido-Merchan, Maria Coronado-Vaca
Deep Reinforcement Learning Deep Reinforcement Learning Agent Portfolio Optimization Model Explainable Deep Reinforcement Learning

April 26, 2024

An Explainable Deep Reinforcement Learning Model for Warfarin Maintenance Dosing Using Policy Distillation and Action Forging
Sadjad Anzabi Zadeh, W. Nick Street, Barrett W. Thomas
Deep Reinforcement Learning Proximal Policy Optimization Action Feature Policy Distillation Dosing Model Explainable Deep Reinforcement Learning Warfarin Dosing

July 18, 2023

IxDRL: A Novel Explainable Deep Reinforcement Learning Toolkit based on Analyses of Interestingness
Pedro Sequeira, Melinda Gervasio
Reinforcement Learning General Analysis User Interest Sequential Decision Making Task RL Agent Explainable Deep Reinforcement Learning

January 24, 2023

Explainable Deep Reinforcement Learning: State of the Art and Challenges
George A. Vouros
Reinforcement Learning Technical Challenge High Explainability Internal State Art Specific Information Explainable Reinforcement Learning Explainability Problem Explainable Deep Reinforcement Learning

June 6, 2022

Robust Adversarial Attacks Detection based on Explainable Deep Reinforcement Learning For UAV Guidance and Planning
Thomas Hickling, Nabil Aouf, Phillippa Spencer
Adversarial Attack Deep Reinforcement Learning Task Planning Obstacle Avoidance Various Fast Moving Drone Adversarial Detection Explainable Deep Reinforcement Learning

November 7, 2021

Explainable Deep Reinforcement Learning for Portfolio Management: An Empirical Approach
Mao Guan, Xiao-Yang Liu
Deep Learning Deep Neural Network Deep Reinforcement Learning Portfolio Management Explainable Deep Reinforcement Learning Prediction Sorted Portfolio

Explainable Deep Reinforcement Learning

Papers

Using Part-based Representations for Explainable Deep Reinforcement Learning

Explainable Post hoc Portfolio Management Financial Policy of a Deep Reinforcement Learning agent

An Explainable Deep Reinforcement Learning Model for Warfarin Maintenance Dosing Using Policy Distillation and Action Forging

IxDRL: A Novel Explainable Deep Reinforcement Learning Toolkit based on Analyses of Interestingness

Explainable Deep Reinforcement Learning: State of the Art and Challenges

Robust Adversarial Attacks Detection based on Explainable Deep Reinforcement Learning For UAV Guidance and Planning

Explainable Deep Reinforcement Learning for Portfolio Management: An Empirical Approach