Admissable: Training Reinforcement Learning Agents against Adversarial Missingness
cs.LG
Submitted: 2026-09-14
Updated: 2026-09-14
License: http://creativecommons.org/licenses/by-sa/4.0/
The gist: In order to make Reinforcement Learning algorithms applicable in real world scenarios, safety must be ensured even under adverse operating conditions.
Terminology
Abstract
In order to make Reinforcement Learning algorithms applicable in real world scenarios, safety must be ensured even under adverse operating conditions. In this work, we consider the challenge of adversarial feature missingness: a scenario in which an adversary occludes features from the agent's observation in order to reduce performance as much as possible. We formally define adversarial missingness for Reinforcement Learning and compare it to the related concepts of infinity-norm bounded adversarial perturbations and learning with missing data. We develop an adversarial training algorithm and show its effectiveness in increasing robustness against adversarial missingness on three MuJoCo benchmark environments. Compared to a baseline trained with random uniform missingness, our method achieves better robustness on all three tasks.
Sources
- On Uncertainty in Deep State Space Models for Model-Based Reinforcement Learning
- Missingness as Stability: Understanding the Structure of Missingness in Longitudinal EHR data and its Impact on Reinforcement Learning in Healthcare
- Learning to Walk in the Real World with Minimal Human Effort
- Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
- Adversarial Attacks on Neural Network Policies
- Policy Smoothing for Provably Robust Reinforcement Learning
- Reinforcement Learning Approaches for Traffic Signal Control under Missing Data
- Robust Deep Reinforcement Learning with Adversarial Attacks
- Intriguing properties of neural networks
- Robust Reinforcement Learning in POMDPs with Incomplete and Noisy Observations
- Robust Reinforcement Learning on State Observations with Learned Optimal Adversary
- Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks