Robust Agent

Robust agent research focuses on developing artificial intelligence agents resilient to unexpected situations, including adversarial attacks and environmental changes, ensuring reliable performance in real-world deployments. Current efforts concentrate on improving agent robustness through techniques like input transformations (e.g., vector quantization), adversarial training, and the development of verifiable robustness guarantees using set-based reinforcement learning and smoothed deep reinforcement learning algorithms. This work is crucial for deploying AI agents in safety-critical applications, such as autonomous driving and robotics, where reliability and adaptability are paramount.

Papers

November 4, 2024

Taking AI Welfare Seriously
Robert Long, Jeff Sebo, Patrick Butlin, Kathleen Finlinson, Kyle Fish, Jacqueline Harding, Jacob Pfau, Toni Sims, Jonathan Birch, David Chalmers
Artificial Intelligence AI System Artificial Intelligence Policy Robust Agent

November 1, 2024

Uncertainty-based Offline Variational Bayesian Reinforcement Learning for Robustness under Diverse Data Corruptions
Rui Yang, Jie Wang, Guoping Wu, Bin Li
Native Robustness Bayesian Inference Robust Offline Reinforcement Learning Bayesian Inference Framework Data Corruption Robust Agent

October 16, 2024

Dual Action Policy for Robust Sim-to-Real Reinforcement Learning
Ng Wen Zheng Terence, Chen Jianda
Reinforcement Learning Domain Adaptation Sim to Real Reinforcement Learning Uncertainty Aware Exploration Robust Agent

October 4, 2024

Mitigating Adversarial Perturbations for Deep Reinforcement Learning via Vector Quantization
Tung M. Luu, Thanh Nguyen, Tee Joshua Tian Jin, Sungwoon Kim, Chang D. Yoo
Reinforcement Learning Adversarial Attack Deep Reinforcement Learning Adversarial Training Adversarial Perturbation Vector Quantization Robust Agent

August 17, 2024

Training Verifiably Robust Agents Using Set-Based Reinforcement Learning
Manuel Wendl, Lukas Koller, Tobias Ladner, Matthias Althoff
Reinforcement Learning Adversarial Attack Action Space Reachability Analysis Robust Reinforcement Learning Set Valued Robust Agent

July 31, 2024

On the Perturbed States for Transformed Input-robust Reinforcement Learning
Tung M. Luu, Haeyong Kang, Tri Ton, Thanh Nguyen, Chang D. Yoo
Reinforcement Learning Robust Reinforcement Learning Robust Agent State Perturbation

June 26, 2024

Breaking the Barrier: Enhanced Utility and Robustness in Smoothed DRL Agents
Chung-En Sun, Sicun Gao, Tsui-Wei Weng
Deep Reinforcement Learning Native Robustness Task Utility Robust Policy Knowledge Barrier Robust Agent

February 16, 2024

Robust agents learn causal world models
Jonathan Richens, Tom Everitt
Causal Inference Causal Model Underlying Causal Robust Agent Causal World Model

October 5, 2023

Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Control
Neehal Tumma, Mathias Lechner, Noel Loo, Ramin Hasani, Daniela Rus
Continuous Time Sparse Connectivity Recurrent Connection Robust Agent

August 21, 2023

Stabilizing Unsupervised Environment Design with a Learned Adversary
Ishita Mediratta, Minqi Jiang, Jack Parker-Holder, Michael Dennis, Eugene Vinitsky, Tim Rocktäschel
Reinforcement Learning Procedural Content Generation Maze Navigation Unsupervised Environment Design Robust Agent Adversary Knowledge

May 7, 2023

A Framework for Characterizing Novel Environment Transformations in General Environments
Matthew Molineaux, Dustin Dannenhauer, Eric Kildebeck
New Framework Scenario Generation New Environment General Environment Robust Agent

April 12, 2023

Facilitating Sim-to-real by Intrinsic Stochasticity of Real-Time Simulation in Reinforcement Learning for Robot Manipulation
Ram Dershan, Amir M. Soufi Enayati, Zengjie Zhang, Dean Richert, Homayoun Najjaran
Reinforcement Learning Real Time Robot Manipulation Domain Randomization Robust Agent

December 6, 2022

What is the Solution for State-Adversarial Multi-Agent Reinforcement Learning?
Songyang Han, Sanbao Su, Sihong He, Shuo Han, Haizhao Yang, Shaofeng Zou, Fei Miao
Multi Agent Reinforcement Learning Solution Path Robust Policy Sub Optimal Robust Agent Adversarial Multi Adversarial Team Game

April 13, 2022

Safer Autonomous Driving in a Stochastic, Partially-Observable Environment by Hierarchical Contingency Planning
Ugo Lecerf, Christelle Yemdji-Tchassi, Pietro Michiardi
Autonomous Vehicle Stochastic Way Stochastic Environment Observable Environment Safe Autonomous Driving Robust Agent Contingency Planning

November 5, 2021

Learning to Cooperate with Unseen Agent via Meta-Reinforcement Learning
Rujikorn Charakorn, Poramate Manoonpong, Nat Dilokthanakul
LeArning Abstract Long Term Cooperation Meta Reinforcement Learning Robust Agent