AI Paper Atlas

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

Mitigating the Effect of Incidental Correlations on Part-based Learning

Gaurav Bhatt, Deepayan Das, Leonid Sigal, Vineeth N. Balasubramanian

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 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
add foreground-background mixture regularization L_mix (ablation)DPViT pretraining · MiniImageNet few-shot learning1-shot accuracy: 61.24 -> 62.81Improved
add foreground-background mixture regularization L_mix (ablation)DPViT pretraining · MiniImageNet few-shot learning5-shot accuracy: 81.12 -> 83.25Improved
add sparse and orthogonal quality regularization L_Q (ablation)DPViT pretraining · MiniImageNet few-shot learning1-shot accuracy: 62.81 -> 62.15Hurt
add sparse and orthogonal quality regularization L_Q (ablation)DPViT pretraining · MiniImageNet part representations||P||_1: 8.73 -> 0.35Improved
add classification-token invariance loss L_cls^inv (ablation)DPViT fine-tuning · ImageNet-9 M-RANDaccuracy: 82.7 -> 87.5Improved
add classification-token invariance loss L_cls^inv (ablation)DPViT fine-tuning · ImageNet-9 BG-GAPBG-GAP: 8.8 -> 5.9Improved
add foreground-code invariance loss L_p^inv (ablation)DPViT fine-tuning with L_cls^inv · ImageNet-9 M-RANDaccuracy: 87.5 -> 87.2Hurt
increase λ_s and λ_o from 0.5 to 2.0 (ablation)DPViT pretraining · MiniImageNet1-shot accuracy: 73.8 -> 72.4Hurt
use foreground masks for only 10% of training samples instead of all samples (ablation)DPViT pretraining · MiniImageNet1-shot accuracy: 73.8 -> 72.9No clear effect
use K=32 instead of K=64 with n_f=K/2 (ablation)DPViT · MiniImageNet1-shot accuracy: 72.9 -> 72.2Hurt
share MSA and MCA attention layers (ablation)DPViT · MiniImageNet1-shot accuracy: 73.8 -> 73.6No clear effect
replace L2 mix loss with L1 (ablation)DPViT pretraining · MiniImageNet1-shot accuracy: 65.12 -> 58.73Hurt

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