DreamTest: World-Model Surrogates for Search-Based Testing of Deep Reinforcement Learning Agents
cs.SE, cs.LG
Submitted: 2026-10-03
Updated: 2026-10-03
Code: https://github.com/eleurent/highway-env
Terminology
Sources
- Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
- Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics
- Rigorous Agent Evaluation: An Adversarial Approach to Uncover Catastrophic Failures
- The Pursuit of Diversity: Multi-Objective Testing of Deep Reinforcement Learning Agents
- World Models
- Dream to Control: Learning Behaviors by Latent Imagination
- Learning Latent Dynamics for Planning from Pixels
- Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
- Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
- Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
- Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
- Hindsight Experience Replay
- When to Trust Your Model: Model-Based Policy Optimization
- On Calibration of Modern Neural Networks
- Search-Based Testing of Reinforcement Learning
- Testing for Fault Diversity in Reinforcement Learning
- The Value Equivalence Principle for Model-Based Reinforcement Learning
- Objective Mismatch in Model-based Reinforcement Learning
- GAIA-1: A Generative World Model for Autonomous Driving
Related papers
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
- IntentCoding: Amplifying User Intent in Code Generation
- Incentives and Outcomes in Bug Bounties