About me

I am a research scientist at Google DeepMind. I was awarded the MIT Innovators Under 35 award in 2023 for my work on discovering new algorithms with machine learning. I obtained my PhD from the signal processing laboratory in EPFL in 2016.

Research interests: I work on AI for Science, and in particular on using Machine Learning to unlock new results in Mathematics and Computer Science. I am also broadly interested in the reliability of Machine Learning systems, and in particular in computer vision.



Selected publications

Mathematical discoveries from program search with large language models
Bernardino Romera-Paredes, Mohammadamin Barekatain, Alexander Novikov, Matej Balog, M. Pawan Kumar, Emilien Dupont, Francisco J. R. Ruiz, Jordan S. Ellenberg, Pengming Wang, Omar Fawzi, Pushmeet Kohli, Alhussein Fawzi
Nature 2023
Blog post
Press: The Guardian, New Scientist, MIT Technology Review, Nature News

Discovering faster matrix multiplication algorithms with reinforcement learning
Alhussein Fawzi, Matej Balog, Aja Huang, Thomas Hubert, Bernardino Romera-Paredes, Mohammadamin Barekatain, Alexander Novikov, Francisco J. R. Ruiz, Julian Schrittwieser, Grzegorz Swirszcz, David Silver, Demis Hassabis, Pushmeet Kohli
Nature 2022 [Cover]

Blog post
Nature research briefing

Press: MIT Technology Review, New Scientist, The Independent, Venture Beat, Nature

Learning dynamic polynomial proofs
Alhussein Fawzi, Mateusz Malinowski, Hamza Fawzi, Omar Fawzi
Neural Information Processing Systems (NeurIPS) 2019 [Spotlight presentation]

Are labels required for improving adversarial robustness?
Jonathan Uesato, Jean-Baptiste Alayrac, Po-Sen Huang, Robert Stanforth, Alhussein Fawzi, Pushmeet Kohli
Neural Information Processing Systems (NeurIPS) 2019

Adversarial Robustness through Local Linearization
Chongli Qin, James Martens, Sven Gowal, Dilip Krishnan, Dj Dvijotham, Alhussein Fawzi, Soham De, Robert Stanforth, Pushmeet Kohli
Neural Information Processing Systems (NeurIPS) 2019

Robustness via curvature regularization, and vice versa
Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Jonathan Uesato, Pascal Frossard
IEEE Computer Vision and Patter Recognition (CVPR), 2019.

Universal adversarial perturbations
Seyed-Mohsen Moosavi-Dezfooli*, Alhussein Fawzi*, Omar Fawzi, Pascal Frossard (*: Equal contribution)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, 2017. [Oral presentation]

Check the demo on YouTube!
Code available on GitHub

Robustness of classifiers: from adversarial to random noise
Alhussein Fawzi*, Seyed-Mohsen Moosavi-Dezfooli*, Pascal Frossard (*: Equal Contribution)
Neural Information Processing Systems (NIPS), 2016.

DeepFool: a simple and accurate method to fool deep neural networks
Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Pascal Frossard
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA, 2016.


Universal adversarial perturbations

Implementation of the algorithm for computing universal perturbations
GitHub page


Implementation of the DeepFool algorithm for fooling deep neural networks
GitHub page


MATLAB and C++ (with OpenCV) implementations of Manitest to compute the invariance of classifiers to geometric transformations.
See project webpage for download and more information.

Learning Algorithm for Soft-Thresholding (LAST)

Implementation of the DC-based dictionary learning algorithm for soft-thresholding based based classifiers.
Download MATLAB code.