Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Kazuki Osawa

Understanding Approximate Fisher Information for Fast Convergence of Natural Gradient Descent in Wide Neural Networks

Oct 23, 2020
Ryo Karakida, Kazuki Osawa

* NeurIPS 2020 

  Access Paper or Ask Questions

Scalable and Practical Natural Gradient for Large-Scale Deep Learning

Feb 13, 2020
Kazuki Osawa, Yohei Tsuji, Yuichiro Ueno, Akira Naruse, Chuan-Sheng Foo, Rio Yokota

* arXiv admin note: text overlap with arXiv:1811.12019 

  Access Paper or Ask Questions

Practical Deep Learning with Bayesian Principles

Jun 06, 2019
Kazuki Osawa, Siddharth Swaroop, Anirudh Jain, Runa Eschenhagen, Richard E. Turner, Rio Yokota, Mohammad Emtiyaz Khan

* Under review 

  Access Paper or Ask Questions

Second-order Optimization Method for Large Mini-batch: Training ResNet-50 on ImageNet in 35 Epochs

Dec 05, 2018
Kazuki Osawa, Yohei Tsuji, Yuichiro Ueno, Akira Naruse, Rio Yokota, Satoshi Matsuoka

* 10 pages, 7 figures 

  Access Paper or Ask Questions