# Validation of Image-Based Neural Network Controllers through Adaptive Stress Testing

arXiv: https://arxiv.org/abs/2003.02381 · HTML: https://mlsanity.com/paper/2003.02381/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Mar 2020.

## Built on by (most cited first, 19 of 19)

- [How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Review](https://mlsanity.com/paper/2107.12045/) (2021)
- [VeriX: Towards Verified Explainability of Deep Neural Networks](https://mlsanity.com/paper/2212.01051/) (2022)
- [Verification of Image-based Neural Network Controllers Using Generative Models](https://mlsanity.com/paper/2105.07091/) (2021)
- [Efficient Neural Network Analysis with Sum-of-Infeasibilities](https://mlsanity.com/paper/2203.11201/) (2022)
- [How Do We Fail? Stress Testing Perception in Autonomous Vehicles](https://mlsanity.com/paper/2203.14155/) (2022)
- [SAFE-GIL: SAFEty Guided Imitation Learning for Robotic Systems](https://mlsanity.com/paper/2404.05249/) (2024)
- [Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations](https://mlsanity.com/paper/2506.08505/) (2025)
- [MDPFuzz: Testing Models Solving Markov Decision Processes](https://mlsanity.com/paper/2112.02807/) (2021)
- [Formal Mechanistic Interpretability: Automated Circuit Discovery with Provable Guarantees](https://mlsanity.com/paper/2602.16823/) (2026)
- [Testing Rare Downstream Safety Violations via Upstream Adaptive Sampling of Perception Error Models](https://mlsanity.com/paper/2209.09674/) (2022)
- [Risk-Driven Design of Perception Systems](https://mlsanity.com/paper/2205.10677/) (2022)
- [Efficiently Computing Compact Formal Explanations](https://mlsanity.com/paper/2409.03060/) (2024)
- [An Overview of Structural Coverage Metrics for Testing Neural Networks](https://mlsanity.com/paper/2208.03407/) (2022)
- [ZoPE: A Fast Optimizer for ReLU Networks with Low-Dimensional Inputs](https://mlsanity.com/paper/2106.05325/) (2021)
- [A Deep Reinforcement Learning Approach to Rare Event Estimation](https://mlsanity.com/paper/2211.12470/) (2022)
- [Exploring Critical Testing Scenarios for Decision-Making Policies: An LLM Approach](https://mlsanity.com/paper/2412.06684/) (2024)
- [Scalable and Interpretable Verification of Image-based Neural Network Controllers for Autonomous Vehicles](https://mlsanity.com/paper/2501.14009/) (2025)
- [Characterizing the Robustness of Black-Box LLM Planners Under Perturbed Observations with Adaptive Stress Testing](https://mlsanity.com/paper/2505.05665/) (2025)
- [Towards Verified and Targeted Explanations through Formal Methods](https://mlsanity.com/paper/2604.14209/) (2026)

## Abstract (arXiv)

Neural networks have become state-of-the-art for computer vision problems because of their ability to efficiently model complex functions from large amounts of data. While neural networks can be shown to perform well empirically for a variety of tasks, their performance is difficult to guarantee. Neural network verification tools have been developed that can certify robustness with respect to a given input image; however, for neural network systems used in closed-loop controllers, robustness with respect to individual images does not address multi-step properties of the neural network controller and its environment. Furthermore, neural network systems interacting in the physical world and using natural images are operating in a black-box environment, making formal verification intractable. This work combines the adaptive stress testing (AST) framework with neural network verification tools to search for the most likely sequence of image disturbances that cause the neural network controlled system to reach a failure. An autonomous aircraft taxi application is presented, and results show that the AST method finds failures with more likely image disturbances than baseline methods. Further analysis of AST results revealed an explainable cause of the failure, giving insight into the problematic scenarios that should be addressed.