AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Learning to Obstruct Few-Shot Image Classification over Restricted Classes

Amber Yijia Zheng, Chiao-An Yang, Raymond A. Yeh

ECCV 2024 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Embedding adaptation for few-shot classification · 2 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 LTO backbone (poor initialization) before CE few-shot learning on CIFAR100CLIP-RN50 + CE · CIFAR100 (avg over superclasses)DropRatio@2: 1.85 -> 10.76Improved
include the FSC learner in the obstruction objective (LTO vs No F) (ablation)LTO · CIFAR100 CEDropRatio@2: 5.97 -> 10.76Improved
include the FSC learner in the obstruction objective (LTO vs No F) on CoOp (ablation)LTO · CIFAR100 CoOpDropRatio@2: 4.54 -> 4.80No clear effect
maximize restricted-class loss only (Only R) vs LTO (ablation)LTO · SUN397 MetaOptNet 5-shotDropRatio@2: 18.01 -> 1.17Hurt
use LTO instead of machine unlearning (SSD) to obstructCLIP-RN50 + CE · CIFAR100 first 9 superclassesDropRatio@2: 1.09 -> 11.99Improved
give the evaluation few-shot learner 4x more data (ablation)LTO-obstructed CLIP-RN50 + CE · ImageNet superclass 1DropRatio@2: 9.93 -> 2.82Hurt
train the evaluation few-shot learner 4x more epochs (ablation)LTO-obstructed CLIP-RN50 + Tip-Adapter · ImageNet superclass 1DropRatio@2: 5.92 -> 3.16Hurt
evaluate with a different few-shot method than the one used in LTO (ablation)LTO · ImageNet superclass 1DropRatio@2: Tip-Adapter LTO: 5.92 on TA, 4.16 on CE, 7.75 on CoOpMixed
increase SSD training data from 5-shot to 20-shot (ablation)SSD unlearning baseline · CIFAR100DropRatio@2: 0.67 -> 2.22 (CE, superclass 1)Improved
apply LTO to CLIP attribute learning on CelebACLIP-RN50 + CE · CelebA attributesAUROC drop ratio: 8.04 average; Pale_Skin 31.64, Gray_Hair 0.08Improved

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