AI Paper Atlas

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

How to Select Which Active Learning Strategy is Best Suited for Your Specific Problem and Budget

Guy Hacohen, Daphna Weinshall

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Deep active learning · 4 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
dynamically choose low-, random-, or high-budget query strategies using the removal-based derivative testactive learning with ResNet-18 · CIFAR-10, CIFAR-100, and ImageNet-50 across low, medium, and high budgetsaccuracyImproved
use SelectAL rather than the best single query strategyactive learning · CIFAR-10, 100+100; CIFAR-100, 100+100; ImageNet-50, 100+100accuracy: 35.1 ± 0.3; 8.2 ± 0.1; 11.4 ± 0.6No clear effect
use SelectAL rather than the best single query strategyactive learning · CIFAR-10, 25k+5k; CIFAR-100, 30k+7k; ImageNet-50, 25k+5kaccuracy: 88.1 ± 0.1; 61.9 ± 0.1; 80.8 ± 0.2No clear effect
choose strategies dynamically instead of using ProbCover aloneactive learning · CIFAR-10 and CIFAR-100, high budgetsaccuracy: 87.1 ± 0.1 -> 88.1 ± 0.1; 61.4 ± 0.3 -> 61.9 ± 0.1Improved
use BADGE as the high-budget removal-test strategy instead of inverse-TypiClust (ablation)SelectAL's budget-regime test · CIFAR-10accuracy-based transition detectionHurt
allow the regime-test strategy to use the labeled set (ablation)SelectAL's budget-regime test · CIFAR-10accuracy-based transition detectionHurt
use MoCo, SimCLR, or SCAN features in the removal test (ablation)SelectAL · CIFAR-10accuracy-based strategy selectionNo clear effect
replace the pixel representation with MoCo, DINO, SimCLR, or SWAV features (ablation)TypiClust · TinyImageNet, low-budget regimetest accuracyImproved
use an autoencoder representation instead of the other tested feature spaces (ablation)TypiClust · TinyImageNet, low-budget regimetest accuracyHurt
use ProbCover for low budgets and BADGE for high budgets rather than a single strategy across budgetsactive learning · CIFAR-10, CIFAR-100, and ImageNet-50accuracyMixed

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