|Adversarial risk and the dangers of evaluating against weak attacks|
J Uesato, B O'Donoghue, A Oord, P Kohli
ICML 2018, 2018
|Robustfill: Neural program learning under noisy I/O|
J Devlin, J Uesato, S Bhupatiraju, R Singh, A Mohamed, P Kohli
Proceedings of the 34th International Conference on Machine Learning-Volume …, 2017
|On the effectiveness of interval bound propagation for training verifiably robust models|
S Gowal, K Dvijotham, R Stanforth, R Bunel, C Qin, J Uesato, ...
arXiv preprint arXiv:1810.12715, 2018
|Technical report on the cleverhans v2. 1.0 adversarial examples library|
N Papernot, F Faghri, N Carlini, I Goodfellow, R Feinman, A Kurakin, ...
arXiv preprint arXiv:1610.00768, 2016
|Are Labels Required for Improving Adversarial Robustness?|
J Uesato, JB Alayrac, PS Huang, R Stanforth, A Fawzi, P Kohli
NeurIPS 2019, 2019
|Robustness via curvature regularization, and vice versa|
SM Moosavi-Dezfooli, A Fawzi, J Uesato, P Frossard
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern …, 2019
|Scaling language models: Methods, analysis & insights from training gopher|
JW Rae, S Borgeaud, T Cai, K Millican, J Hoffmann, F Song, J Aslanides, ...
arXiv preprint arXiv:2112.11446, 2021
|Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples|
S Gowal, C Qin, J Uesato, T Mann, P Kohli
arXiv preprint arXiv:2010.03593, 2020
|Training verified learners with learned verifiers|
K Dvijotham, S Gowal, R Stanforth, R Arandjelovic, B O'Donoghue, ...
arXiv preprint arXiv:1805.10265, 2018
|Ethical and social risks of harm from Language Models|
L Weidinger, J Mellor, M Rauh, C Griffin, J Uesato, PS Huang, M Cheng, ...
arXiv preprint arXiv:2112.04359, 2021
|Scalable verified training for provably robust image classification|
S Gowal, KD Dvijotham, R Stanforth, R Bunel, C Qin, J Uesato, ...
Proceedings of the IEEE/CVF International Conference on Computer Vision …, 2019
|Rigorous agent evaluation: An adversarial approach to uncover catastrophic failures|
J Uesato, A Kumar, C Szepesvari, T Erez, A Ruderman, K Anderson, ...
ICLR 2019, 2018
|Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming|
S Dathathri, K Dvijotham, A Kurakin, A Raghunathan, J Uesato, RR Bunel, ...
Advances in Neural Information Processing Systems 33, 5318-5331, 2020
|An Alternative Surrogate Loss for PGD-based Adversarial Testing|
S Gowal, J Uesato, C Qin, PS Huang, T Mann, P Kohli
arXiv preprint arXiv:1910.09338, 2019
|Challenges in detoxifying language models|
J Welbl, A Glaese, J Uesato, S Dathathri, J Mellor, LA Hendricks, ...
arXiv preprint arXiv:2109.07445, 2021
|Semantic code repair using neuro-symbolic transformation networks|
J Devlin, J Uesato, R Singh, P Kohli
arXiv preprint arXiv:1710.11054, 2017
|Verification of non-linear specifications for neural networks|
C Qin, B O'Donoghue, R Bunel, R Stanforth, S Gowal, J Uesato, ...
ICLR 2019, 2019
|Uncovering Surprising Behaviors in Reinforcement Learning via Worst-case Analysis|
A Ruderman, R Everett, B Sikder, H Soyer, J Uesato, A Kumar, C Beattie, ...
|Make Sure You're Unsure: A Framework for Verifying Probabilistic Specifications|
L Berrada, S Dathathri, K Dvijotham, R Stanforth, RR Bunel, J Uesato, ...
Advances in Neural Information Processing Systems 34, 2021
|REALab: An Embedded Perspective on Tampering|
R Kumar, J Uesato, R Ngo, T Everitt, V Krakovna, S Legg
arXiv preprint arXiv:2011.08820, 2020