AI Paper Atlas

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

Fast and Accurate Transferability Measurement by Evaluating Intra-class Feature Variance

Huiwen Xu, U Kang

ICCV 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 9 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
measure transferability by intra-class feature variance (TMI) instead of discriminability scorestransferability measurement, supervised models · 17 datasets, Kendall correlationKendall correlation: best in 13 of 17 datasetsMixed
use TMI for top-5 architecture selection rather than TransRatetransferability measurement · 17 target taskstop-5 hit ratio: 9 / 17 -> 14 / 17Improved
use TMI for best source data selection rather than TransRatetransferability measurement · ResNet-50 source data selectiontop-1 hit ratio: 13 / 17 -> 17 / 17Improved
use conditional entropy as the intra-class variance measure instead of Contrast, Center, SNCA or MS terms (ablation)TMI · 17 datasets, Kendall correlationKendall correlation: best in 15 of 17 datasetsImproved
apply TMI to self-supervised pre-trained models without classifiersTMI · 11 self-supervised models, 17 datasetsKendall correlation: best in 13 of 17; VisDA -0.597Mixed
apply TMI to transferring layer selectionTMI · ResNet-18, 17 datasetsKendall correlation: best in 16 of 17 datasetsImproved
use TMI vs baselines to check consistency of transferability sign with accuracy gain from pre-trainingTMI · ImageNet-pretrained vs from-scratch ResNet-50sign consistency: 14 of 17 cases consistentImproved
vary number of neighbors n_k in the entropy estimator (ablation)TMI · hyperparameter sensitivitycorrelationNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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