AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

T-vMF Similarity for Regularizing Intra-Class Feature Distribution

Takumi Kobayashi

CVPR 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 25 citations (OpenAlex)

Check 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
replace cosine similarity with t-vMF similarity (κ=64)ResNet-10 · ImageNet-LTtop-1 error (%): 59.17 -> 56.31Improved
replace cosine similarity with t-vMF similarity (κ=16)ResNet-50 · iNaturalist2018top-1 error (%): 32.20 -> 28.92Improved
replace cosine similarity with t-vMF similarity (κ=64)ResNet-50 · iNaturalist2019top-1 error (%): 26.34 -> 25.08Improved
replace cosine similarity with t-vMF similarity (κ=16)ResNet-10 · ImageNet-Stop-1 error (%): 52.69 -> 52.06Improved
replace cosine similarity with t-vMF similarity (κ=16)ResNet-10 · ImageNet-SStop-1 error (%): 66.41 -> 64.77Improved
replace cosine similarity with t-vMF similarity (κ=64)ResNet-10 · ImageNet-Ntop-1 error (%): 72.40 -> 69.19Improved
use t-vMF with κ=-0.3 rather than ArcFaceResNet-50 · ImageNettop-1 error (%): 23.28 -> 22.62Improved
increase κ from 4 to 64 (ablation)t-vMF on ResNet-10 · ImageNet-LTtop-1 error (%): 59.17 -> 56.31Improved
replace t-vMF with vMF similarity at κ=16 (ablation)ResNet-10 · ImageNet-LTtop-1 error (%): 57.30 -> 75.84Hurt
replace t-vMF with q-vMF similarity (ablation)ResNet-10 · ImageNet-LTtop-1 error (%): 57.30 -> 59.05Hurt
set κ=0 to recover cosine similarity (ablation)t-vMF on ResNet-10 · ImageNet-LTtop-1 error (%): 56.31 -> 61.32Hurt
learn a separate κ per class rather than fixing κ (ablation)t-vMF on ResNet-10 · ImageNet-LTtop-1 error (%): 56.31 -> 60.39Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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