AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Simplifying Momentum-based Positive-definite Submanifold Optimization with Applications to Deep Learning

Wu Lin, Valentin Duruisseaux, Melvin Leok, Frank Nielsen, Mohammad Emtiyaz Khan, Mark Schmidt

ICML 2023 · Learning theory and algorithms › Matrix and tensor methods · approach: Randomized sketching for matrices · 0 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
Use the proposed optimizer instead of SGDVGG-16 · CIFAR-100test error rate: 26.45 -> 26.06Improved
Use the proposed optimizer instead of SGDPyramidNet-65 · CIFAR-100test error rate: 25.65 -> 25.39Improved
Use the proposed optimizer instead of KFACConvMixer256-12 · CIFAR-100test error rate: 26.14 -> 25.79Improved
Use the proposed optimizer instead of SGDRegNetX-1.6GF · CIFAR-100test error rate: 21.62 -> 21.89Hurt
Use the proposed optimizer instead of SGDVGG-16 · TinyImageNet-200test error rate: 40.18 -> 40.01Improved
Use the proposed optimizer instead of KFACPyramidNet-65 · TinyImageNet-200test error rate: 41.23 -> 40.82Improved
Use the proposed optimizer instead of SGDConvMixer256-12 · TinyImageNet-200test error rate: 40.03 -> 39.09Improved
Use the proposed optimizer instead of SGDRegNetX-1.6GF · TinyImageNet-200test error rate: 34.38 -> 34.85Hurt
Use the proposed optimizer instead of SGDConvMixer180-16 · ImageNet-100test error rate: 20.37 -> 18.84Improved
Use the proposed optimizer instead of KFACRegNetZ-500MF · ImageNet-100test error rate: 20.07 -> 18.30Improved
Use the proposed optimizer instead of SGDReXNetr-100 · ImageNet-100test error rate: 17.07 -> 16.73Improved
Use the proposed optimizer instead of SGDRepVGG-B1G4 · ImageNet-100test error rate: 16.15 -> 15.81Improved

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