
IS
I. Samuel Akinwande, Mykel J. Kochenderfer, Clark Barrett
· 1 min read
ResearcharXiv cs.AI
Stochastic World Models for Verifying Vision-Based Neural Feedback Systems
arXiv:2609.38120v1 Announce Type: new
Abstract: Verifying a vision-based neural feedback system requires a model of the observations its controller acts upon. Such a model must capture the variation the sensor produces, while remaining tractable for closed-loop analysis. Generative adversarial networks (GANs) have served as perception surrogates, but they are large, reproduce complex scenes poorly, and are hard to verify. We explore stochastic world models as a richer class of perception surrogates. We train a world model with physically grounded latents, built from operations that standard verifiers bound. It reproduces held-out frames more faithfully than GAN surrogates with up to 130 times as many parameters. To verify these surrogates, we develop a procedure that combines falsification, adaptive refinement, symbolic, and backward analyses. On an emergency braking benchmark with a GAN surrogate, our procedure resolves the entire state space, 38% of which the state-of-the-art verifier left unresolved. On the RGB version of the benchmark, where no verification results have previously been reported, our procedure resolves over 80% of the state space with a world model surrogate.
Original source
This story was published by arXiv cs.AI and written by I. Samuel Akinwande, Mykel J. Kochenderfer, Clark Barrett. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


