AI Paper Atlas

Research map / Optimization / Optimizers

Scalable and Practical Natural Gradient for Large-Scale Deep Learning

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

TPAMI 2020 · Optimization › Optimizers · approach: Memory-efficient LLM optimizers · 44 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add stale statistics (adaptive update frequency) (ablation)SP-NGD emp+unitBN, BS=4K · ImageNet ResNet-50speedup / top-1: 75.2 % ->74.8 %, x 1.33 speedup, 23.6 % communicationMixed
add stale statistics (ablation)SP-NGD, BS=16K · ImageNet ResNet-50speedup: x 1.68 speedup, 5.4 % communication, 75.3 % ->75.2 %Improved
add stale statistics (ablation)SP-NGD, BS=32K · ImageNet ResNet-50speedup: x 1.40 speedup, 7.8 % communication, 75.6 % ->75.4 %Improved
use empirical Fisher instead of single Monte Carlo Fisher estimate (ablation)SP-NGD · ImageNet ResNet-50time per stepImproved
use unit-wise NGD on BatchNorm instead of full BN Fisher (ablation)SP-NGD · ImageNet ResNet-50time per stepImproved
increase batch size from 32K to 131KSP-NGD · ImageNet ResNet-50top-1 accuracy: 75.4 % ->74.9 %Mixed
use second-order NGD instead of SGD for large batchSP-NGD vs SGD · ImageNet ResNet-50 BS=32Kepochs: 45 epochs vs 90 epochsImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy