When to trust AI
Machine learning solutions are revolutionizing AI, but Marta Kwiatkowska explores their instability against adversarial examplessmall perturbations to inputs that can catastrophically affect the outputwhich raises concerns about the readiness of this technology for widespread deployment.
| Talk Title | When to trust AI |
| Speakers | Marta Kwiatkowska (University of Oxford) |
| Conference | O’Reilly Artificial Intelligence Conference |
| Conf Tag | Put AI to Work |
| Location | London, United Kingdom |
| Date | October 15-17, 2019 |
| URL | Talk Page |
| Slides | |
| Video | Talk Video |
Machine learning solutions are revolutionizing AI, but their instability against adversarial examples—small perturbations to inputs that can catastrophically affect the output—raises concerns about the readiness of this technology for widespread deployment. Marta Kwiatkowska uses illustrative examples to give you an overview of techniques being developed to improve the robustness, safety, and trust in AI systems.
