White Paper 2019 (Validate AI)
Jump to navigation Jump to search
The aim of the Validate AI conference was to explore how such systems can depart from this ideal, to examine tools and methods for ensuring sound and appropriate behaviour in a variety of different application domains under different circumstances, and to look at open challenges in validation. Issues explored included accurate and unbiased performance and its evaluation, model testing and formal verification, ensuring resilience against adversarial attacks, and the effective maintenance of systems as their working environment evolves. Representatives from public, private and academic organisations were brought together to give their perspectives on the issues and the way forward, and to share ideas and solutions.