Kazuki Osawa

4 papers · 2019–2023 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🗺️ Taxonomy Completionist (12)

Venues

NeurIPS (2) CVPR (1) ICML (1)

Top co-authors

Rio Yokota (2) Ryo Karakida (2) Tomoumi Takase (1) Siddharth Swaroop (1) Richard E Turner (1) Tomohiro Hayase (1) Anirudh Jain (1) Runa Eschenhagen (1) Akira Naruse (1) Yohei Tsuji (1)

Research topics

Machine Learning (4) Optimization & Theory · Machine Learning (4) Deep Learning (2) Architectures (2) Optimization & Theory · Deep Learning (2) Core Methods (1) Learning Types (1)

Keywords

deep learning (2) neural network optimization (2) variational inference (1) convergence analysis (1) distributed learning (1) fisher information (1) finite difference (1) natural gradient descent (1) distributed training (1) second-order optimization (1) convolutional neural network (1) implicit bia (1) bayesian deep learning (1) flat minima (1) generalization performance (1) diagonal linear network (1) gradient regularization (1) kronecker factorization (1) finite-difference computation (1) kronecker-factored approximate curvature (1)

Papers

Understanding Gradient Regularization in Deep Learning: Efficient Finite-Difference Computation and Implicit Bias ICML 2023 Understanding Approximate Fisher Information for Fast Convergence of Natural Gradient Descent in Wide Neural Networks NeurIPS 2020 Large-Scale Distributed Second-Order Optimization Using Kronecker-Factored Approximate Curvature for Deep Convolutional Neural Networks CVPR 2019 Practical Deep Learning with Bayesian Principles NeurIPS 2019