Towards Verified Artificial Intelligence
Sanjit A. Seshia, Dorsa Sadigh, and S. Shankar Sastry
TL;DR
This paper discusses the importance of verified AI, aiming for systems with mathematically provable correctness, and outlines key challenges and principles from a formal methods perspective.
Contribution
It identifies five major challenges in achieving verified AI and proposes five principles to address these challenges from a formal methods standpoint.
Findings
Five challenges for verified AI identified
Five principles proposed to overcome these challenges
Formal methods are emphasized for ensuring AI correctness
Abstract
Verified artificial intelligence (AI) is the goal of designing AI-based systems that that have strong, ideally provable, assurances of correctness with respect to mathematically-specified requirements. This paper considers Verified AI from a formal methods perspective. We describe five challenges for achieving Verified AI, and five corresponding principles for addressing these challenges.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFormal Methods in Verification · Software Testing and Debugging Techniques · Adversarial Robustness in Machine Learning