AI Paper Atlas

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

RGMIL: Guide Your Multiple-Instance Learning Model with Regressor

Zhaolong Du, Shasha Mao, Yimeng Zhang, Shuiping Gou, Licheng Jiao, Lin Xiong

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification · 1 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 normalization formula to baseline RGP (ablation)RGP · MMNIST 512/1instance-level accuracy (%): 66.54 -> 72.70Improved
add normalization formula to baseline RGP (ablation)RGP · MMNIST 64/1instance-level accuracy (%): 62.75 -> 90.88Improved
add difference formula on top of normalization (ablation)RGP · MMNIST 512/1instance-level accuracy (%): 72.70 -> 87.28Improved
add difference formula on top of normalization (ablation)RGP · MMNIST 64/1instance-level accuracy (%): 90.88 -> 94.50Improved
add difference formula on top of normalization (ablation)RGP · MMNIST 10/1instance-level accuracy (%): 95.31 -> 98.25Improved
add normalization formula to baseline RGP (ablation)RGP · MMNIST 10/1instance-level accuracy (%): 94.60 -> 95.31No clear effect
increase training bag size from 10 to 512RGP · MMNIST instance-levelinstance-level accuracy (%): 98.25 -> 87.28Hurt
increase training bag size from 10 to 512ABP · MMNIST instance-levelinstance-level accuracy (%): 92.18 -> 42.21Hurt
replace RGP with max pooling aggregatorRGMIL · MMNIST 512/1instance-level accuracy (%): 87.28 -> 65.37Hurt
replace RGP with gated attention pooling (G-ABP)RGMIL · MMNIST 10/1instance-level accuracy (%): 98.25 -> 91.47Hurt
use RGP instead of attention pooling (ABP)RGMIL · MMNIST 4-class generalinstance-level accuracy (%): 71.76 -> 88.79Improved
use RGP instead of attention pooling (ABP)RGMIL · MMNIST 4-class general 512/1instance-level accuracy (%): 62.43 -> 81.46Improved

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