Brian McWilliams

AI Research Scientist at DeepMind

I am a research scientist at DeepMind. I am particularly interested in theoretical aspects of deep learning, randomized algorithms, sampling and optimization for large scale learning. I am passionate about the use of AI for conservation. See my Google Scholar profile and papers below for more. Contact me at bmcw @ google.

Until July 2018 I headed the Deep Learning and Analytics group at Disney Research Zurich. I worked on developing ML algorithms for production quality rendering, image and video processing, contributing to Toy Story 4 and Frozen 2. Once, a video I narrated went a bit viral.

Between 2012 and 2015 I was a postdoctoral researcher and lecturer in the Institute of Machine Learning at ETH Zurich.

I completed my Ph.D. in 2012 in the Statistics Section of the Department of Mathematics Imperial College, London. Previously, I did the MSc. in Informatics specialising in Machine Learning, Neuroinformatics and Intelligent Robotics at the University of Edinburgh and a BEng. in Computer Systems Engineering at the University of Warwick.

New.

Social Diversity and Social Preferences in Mixed-Motive Reinforcement Learning

K McKee, E Hughes, J Leibo, I Gemp, B McWilliams, E Duéñez-Guzmán | AAMAS 2020 | arXiv

The Unreasonable Effectiveness of Adam on Cycles

I Gemp, B McWilliams | Bridging Game Theory & Deep Learning 2019 | paper

Using machine learning to accelerate ecological research

"DeepMind is collaborating with ecologists and conservationists to develop machine learning methods to help study the behavioural dynamics of an entire African animal community in the Serengeti National Park and Grumeti Reserve in Tanzania."

S Petersen, et al | blog post

Selected Papers.

Neural Importance Sampling

T Müller, B McWilliams, F Rousselle, M Gross, J Novak | arXiv | video | interactive test suite | SIGGRAPH 2019

Denoising with Kernel Prediction and Asymmetric Loss Functions

T Vogels, F Rousselle, B McWilliams, G Rothlin, A Harvill, D Adler, M Meyer, J Novak | SIGGRAPH 2018 | paper | video | slides

PhaseNet for Video Frame Interpolation

S Meyer, A Djelouah, C Schroers, B McWilliams, A Sorkine-Hornung, M Gross | CVPR 2018 | arXiv

A Fully Progressive Approach to Single-Image Super-Resolution

Y Wang, F Perazzi, B McWilliams, A Sorkine-Hornung, O Sorkine-Hornung, C Schröers | CVPR NTIRE 2018 | arXiv | code | 2 minute summary

Deep Scattering: Rendering Atmospheric Clouds with Radiance-Predicting Neural Networks

S Kallweit, T Muller, B McWilliams, M Gross, J Novak | SIGGRAPH Asia 2017 | arXiv | project page | 2 minute summary

The Shattered Gradients Problem: If resnets are the answer, then what is the question?

D Balduzzi, M Frean, L Leary, JP Lewis, K Ma, B McWilliams | ICML 2017 | arXiv | video

Neural Taylor Approximation: Convergence and Exploration in Rectifier Networks

D Balduzzi, B McWilliams, T Butler-Yeoman | ICML 2017 | arXiv | video

Kernel-predicting Convolutional Networks for Denoising Monte Carlo Renderings

S Bako, T Vogels, B McWilliams, M Meyer, J Novak, A Harvill, P Sen, T DeRose, F Rouselle | SIGGRAPH 2017 | Project page

Preserving Differential Privacy Between Features in Distributed Estimation

C Heinze-Deml, B McWilliams, N Meinshausen | Stat | arXiv

Scalable Adaptive Stochastic Optimization Using Random Projections

G Krummenacher, B McWilliams, Y Kilcher, J Buhmann, N Meinshausen | NIPS 2016 | arXiv

A Benchmark Dataset and Evaluation Methodology for Video Object Segmentation

F Perazzi, J Pont-Tuset, B McWilliams, M Gross, L Van Gool, A Sorkine-Hornung | CVPR 2016 | Project page

DUAL-LOCO: Distributing Statistical Estimation Using Random Projections

C Heinze, B McWilliams, N Meinshausen | AISTATS 2016 | arXiv | software

Variance Reduced Stochastic Gradient Descent with Neighbors

T Hofmann, A Lucchi, S Lacoste-Julien, B McWilliams | NIPS 2015 | arXiv

LOCO: Distributing Ridge Regression with Random Projections.

C Heinze, B McWilliams, N Meinshausen, G Krummenacher | arXiv | software

Fast and Robust Least Squares Estimation in Corrupted Linear Models

B McWilliams, G Krummenacher, M Lučić, J Buhmann | NIPS 2014 | arXiv | Software | slides | video

Correlated random features for fast semi-supervised learning.

B McWilliams, D Balduzzi, J Buhmann | NIPS 2013 | arXiv

More Papers.

Social Diversity and Social Preferences in Mixed-Motive Reinforcement LearningK McKee, E Hughes, J Leibo, I Gemp, B McWilliams, E Duéñez-Guzmán | AAMAS 2020 | arXiv
The Unreasonable Effectiveness of Adam on CyclesI Gemp, B McWilliams | Bridging Game Theory & Deep Learning | paper
Spectrogram Feature Losses for Music Source Separation. A Sahai, R Weber, B McWilliams | EUSIPCO 2019 | arXiv
Automatically Learning an Intuitive Animation Interface From a Collection of Human Motion Clips. M Lüdi, M Guay, B McWilliams, R W Sumner | CGVCVIP 2017.
Learning Representations for Outlier Detection on a Budget. B Micenková, B McWilliams, I Assent | arXiv
A Variance Reduced Stochastic Newton MethodA Lucchi, B McWilliams, T Hofmann | arXiv
RadaGrad: Random Projections for Adaptive Stochastic Optimization. G Krummenacher, B McWilliams | 7th NIPS WS on OptML
Learning Outlier Ensembles: The Best of Both Worlds – Supervised and Unsupervised. B Micenková, B McWilliams, I Assent | KDD WS on ODD2
Subspace clustering of high-dimensional data: a predictive approach. B McWilliams, G Montana | Data Mining and Knowledge Discovery. 28(3): 736-772 | arXiv
Pruning random features with correlated kitchen sinks B McWilliams, D Balduzzi | SPARS 2013.
Projection based models for high dimensional data. Ph.D. thesis.
Multi-view predictive partitioning in high dimensions. B McWilliams, G Montana | Statistical Analysis and Data Mining (2012). 5: 304-321. arXiv.
Predictive subspace clustering. B McWilliams, G Montana. In Proceedings of the 10th International Conference on Machine Learning and Applications (2011), 247-252 | paper
A PRESS statistic for two-block partial least squares regression. B McWilliams, G Montana | In Proceedings of the 10th Conference on Computational Intelligence UK (2010), Colchester | paper
Sparse partial least squares for on-line variable selection in multivariate data streams. B McWilliams, G Montana | Statistical Analysis and Data Mining (2010). 3: 170-193
Predictive modeling with high-dimensional data streams: an on-line variable selection approach. B McWilliams, G Montana | SPARS 2009.