undefined

Matjaz Leonardis

Advancing the understanding of a theorem by Popper and Miller from 1983, offering new insights into Bayesian updating and its standard interpretations, with research interests in learning and creativity in humans and machines.

Best podcasts with Matjaz Leonardis

Ranked by the Snipd community
undefined
Dec 15, 2023 • 9min

Matjaz Leonardis | Interpretability and Security of AI Models

Matjaz Leonardis discusses the risks of undetectable back doors in ML models, their impact on model integrity, challenges to interpretability and robustness, and the need for deeper research into vulnerabilities.