AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

General Greedy De-Bias Learning

Xinzhe Han, Shuhui Wang, Chi Su, Qingming Huang, Qi Chuan Tian

TPAMI 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Out-of-distribution generalization theory · 30 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
use curriculum regularization (GGD_cr) instead of constant greedy removal (GGD_gs) (ablation)GGD dq · VQA-CP v2 testaccuracy: 57.12 -> 59.57Improved
add distribution bias to shortcut bias (ablation)GGD_cr · VQA-CP v2 testaccuracy: 55.81 -> 59.57Improved
add shortcut bias to distribution bias (ablation)GGD_gs · VQA-CP v2 testaccuracy: 48.27 -> 57.12Improved
use greedy learning instead of summing biased and base outputs (ablation)UpDn · VQA-CP v2 testaccuracy: 35.46 -> 57.12Improved
use greedy GGD_cr instead of LMH+RUBi combination (ablation)UpDn · VQA-CP v2 testaccuracy: 51.54 -> 59.57Improved
relax lambda_t from 1.0 to 0.95 (ablation)GGD_gs · Biased-MNIST (rho_train 0.999)accuracy: in-distribution 91.42 -> 99.96; OOD 68.31 -> 58.79Mixed
anneal lambda_t (curriculum) instead of constant 1 (ablation)GGD_gs · Biased-MNIST (rho_train 0.999)accuracy: in-distribution 91.42 -> 99.58; OOD 68.31 -> 67.01Improved
add self-ensemble GGD_cr without predefined bias (ablation)UpDn · VQA-CP v2 testaccuracy: 39.89 -> 54.42Improved
remove distribution bias before self-ensemble (ablation)GGD_cr^se · VQA-CP v2 testaccuracy: 54.42 -> 57.08Improved
use GGD_cr^1k instead of ReBias^1k (ablation)ResNet-18 · Biased-MNIST (rho_train 0.999, rho_test 0)accuracy: 32.84 -> 68.36Improved
use constant greedy GGD_gs dq on in-distribution dataUpDn · VQA v2 valaccuracy: 63.79 -> 59.30Hurt
add GGD_cr with extra AddSentOne hard examplesBiDAF · Adversarial SQuAD AddSentF1: 50.11 -> 53.94Improved

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