seshiaVerifiedArtificialIntelligence2020:
  type: article
  title: Towards {Verified Artificial Intelligence}
  author:
  - Seshia, Sanjit A.
  - Sadigh, Dorsa
  - Sastry, S. Shankar
  date: 2020-07-23
  url:
    value: http://arxiv.org/abs/1606.08514
    date: 2025-02-19
  serial-number:
    arxiv: '1606.08514'
    doi: 10.48550/arXiv.1606.08514
  language: en-US
  abstract: Verified artificial intelligence (AI) is the goal of designing AI-based systems 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.
  parent:
    type: periodical