Timezone: »
Prof. Marta Kwiatkowska (Oxford): Robustness Guarantees for Bayesian Neural Networks
Marta Kwiatkowska
Fri Jul 28 12:10 PM -- 12:40 PM (PDT) @
Bayesian neural networks (BNNs), a family of neural networks with a probability distribution placed on their weights, have the advantage of being able to reason about uncertainty in their predictions as well as data. Their deployment in safety-critical applications demands rigorous robustness guarantees. This paper summarises recent progress in developing algorithmic methods to ensure certifiable safety and robustness guarantees for BNNs, with the view to support design automation for systems incorporating BNN components.
Author Information
Marta Kwiatkowska (University of Oxford)
More from the Same Authors
-
2022 Poster: Learning Dynamics and Generalization in Deep Reinforcement Learning »
Clare Lyle · Mark Rowland · Will Dabney · Marta Kwiatkowska · Yarin Gal -
2022 Spotlight: Learning Dynamics and Generalization in Deep Reinforcement Learning »
Clare Lyle · Mark Rowland · Will Dabney · Marta Kwiatkowska · Yarin Gal -
2022 Poster: Tractable Uncertainty for Structure Learning »
Benjie Wang · Matthew Wicker · Marta Kwiatkowska -
2022 Oral: Tractable Uncertainty for Structure Learning »
Benjie Wang · Matthew Wicker · Marta Kwiatkowska -
2020 Poster: Invariant Causal Prediction for Block MDPs »
Amy Zhang · Clare Lyle · Shagun Sodhani · Angelos Filos · Marta Kwiatkowska · Joelle Pineau · Yarin Gal · Doina Precup