Safety

Publication

Fair Normalizing Flows

Mislav Balunovic *, Anian Ruoss, Martin Vechev *

ICLR

Download

Publication

Your Policy Regularizer is Secretly an Adversary

Rob Brekelmans *, Tim Genewein, Jordi Grau, Grégoire Delétang, Markus Kunesch, Shane Legg, Pedro Ortega

arXiv

Download

Publication

Defending Against Image Corruptions Through Adversarial Augmentations

Dan A. Calian, Florian Stimberg, Olivia Wiles, Sylvestre-Alvise Rebuffi, Andras Gyorgy, Timothy Mann, Sven Gowal

arXiv

Download

Publication

Equilibrium Refinements for Multi-Agent Influence Diagrams: Theory and Practice

James Fox *, Lewis Hammond *, Tom Everitt, Alessandro Abate *, Michael Wooldridge *

AAMAS

Download

Publication

Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming

Dj Dvijotham, Jonathan Uesato, Sumanth Dathathri, Rudy Bunel, Pushmeet Kohli

NeurIPS

Download

Publication

Avoiding Side Effects By Considering Future Tasks

Victoria Krakovna, Laurent Orseau, Richard Ngo, Miljan Martic, Shane Legg

NeurIPS

Download

Publication

Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples

Sven Gowal, Chongli Qin, Jonathan Uesato, Timothy Mann, Pushmeet Kohli

arXiv

Download

Publication

Pessimism About Unknown Unknowns Inspires Conservatism

Michael Cohen *, Marcus Hutter

arXiv

Download

Publication

Efficient and Scalable Bayesian Neural Nets with Rank-1 Factors

Mike Dusenberry *, G Jerfel *, Y Wen *, Y Ma *, Jasper Snoek *, Katherine Heller *, Balaji Lakshminarayanan, D Tran *

ICML

Download

Publication

Pitfalls of learning a reward function online

Stuart Armstrong *, Jan Leike, Laurent Orseau, Shane Legg

arXiv

Download

Publication

An empirical investigation of the challenges of real-world reinforcement learning

Gabriel Dulac-Arnold *, Nir Levine, Daniel Mankowitz, Jerry Li, Cosmin Paduraru, Sven Gowal, Todd Hester

arXiv

Download

Publication

The Incentives that Shape Behaviour

Ryan Carey *, Eric Langlois, Tom Everitt, Shane Legg

arXiv

Download

Publication

Artificial Intelligence, Values and Alignment

Iason Gabriel

arXiv

Download

Publication

Deep Ensembles: A Loss Landscape Perspective

Stanislav Fort *, Huiyi Hu *, Balaji Lakshminarayanan

arXiv

Download

Publication

AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty

Dan Hendrycks, Norman Mu *, E Cubuk *, Barret Zoph *, Justin Gilmer *, Balaji Lakshminarayanan

arXiv

Download

Publication

Towards Robust Image Classification Using Sequential Attention Models

Daniel Zoran, M Chrzanowski *, Po-Sen Huang, Sven Gowal, Alex Mott, Pushmeet Kohli

arXiv

Download

Publication

An Alternative Surrogate Loss for PGD-based Adversarial Testing

Sven Gowal, Jonathan Uesato, Chongli Qin, Po-Sen Huang, Timothy Mann, Pushmeet Kohli

arXiv

Download

Publication

Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation

Po-Sen Huang, Robert Stanforth, Johannes Welbl *, Chris Dyer, Dani Yogatama, Sven Gowal, Krishnamurthy (Dj) Dvijotham, Pushmeet Kohli

EMNLP

Download

Publication

Reward Tampering Problems and Solutions in Reinforcement Learning: A Causal Influence Diagram Perspective

Tom Everitt, Marcus Hutter

arXiv

Download

Publication

Adversarial Robustness through Local Linearization

Chongli Qin, James Martens, Sven Gowal, Dilip Krishnan *, Krishnamurthy (Dj) Dvijotham, Alhussein Fawzi, Soham De, Robert Stanforth, Pushmeet Kohli

arXiv

Download

Publication

Modeling AGI Safety Frameworks with Causal Influence Diagrams

Tom Everitt, Ramana Kumar, Victoria Krakovna, Shane Legg

IJCAI

Download

Publication

Likelihood Ratios for Out-of-Distribution Detection

J Ren *, P Liu *, Emily Fertig *, Jasper Snoek *, Ryan Poplin *, M DePristo *, J Dillon *, Balaji Lakshminarayanan

arXiv

Download

Publication

Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift

Yaniv Ovadia *, Emily Fertig *, Z Nado *, N Nowozin *, J Dillon *, Balaji Lakshminarayanan, Jasper Snoek *

arXiv

Download

Publication

Training verified learners with learned verifiers

Krishnamurthy (Dj) Dvijotham, Sven Gowal, Robert Stanforth, Relja Arandjelovic, Brendan O'Donoghue, Jonathan Uesato, Pushmeet Kohli

arXiv

Download

Publication

Knowing When to Stop: Evaluation and Verification of Conformity to Output-size Specifications

Chenglong Wang, Rudy Bunel *, Krishnamurthy (Dj) Dvijotham, Po-Sen Huang, Edward Grefenstette, Pushmeet Kohli

arXiv

Download

Publications

Safety