I am a PhD student in the Machine Learning Group within the Computational and Biological Learning Lab at the University of Cambridge, supervised by Richard Turner and advised by Carl Edward Rasmussen. I am broadly interested in better understanding and improving deep learning. Specifically, I work on approximate Bayesian inference and training algorithms for deep learning.

Previously, I obtained an MSc in Machine Learning from the University of Tübingen and worked as a research assistant in the Methods of Machine Learning group led by Philipp Hennig. I received a BSc in Cognitive Science from the University of Osnabrück and spent my final year as an intern in the Approximate Bayesian Inference Team led by Emtiyaz Khan at RIKEN AIP in Tokyo.

See my personal website for more details.

Publications

Practical Deep Learning with Bayesian Principles

Kazuki Osawa, Siddharth Swaroop, Anirudh Jain, Runa Eschenhagen, Richard E. Turner, Rio Yokota, Mohammad Emtiyaz Khan, 2019. (In Advances in Neural Information Processing Systems 33).

Abstract▼ URL

Bayesian methods promise to fix many shortcomings of deep learning, but they are impractical and rarely match the performance of standard methods, let alone improve them. In this paper, we demonstrate practical training of deep networks with natural-gradient variational inference. By applying techniques such as batch normalisation, data augmentation, and distributed training, we achieve similar performance in about the same number of epochs as the Adam optimiser, even on large datasets such as ImageNet. Importantly, the benefits of Bayesian principles are preserved: predictive probabilities are well-calibrated, uncertainties on out-of-distribution data are improved, and continual-learning performance is boosted. This work enables practical deep learning while preserving benefits of Bayesian principles. A PyTorch implementation is available as a plug-and-play optimiser.

No matching items
Back to top