AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

ORIENT: Submodular Mutual Information Measures for Data Subset Selection under Distribution Shift

Athresh Karanam, Krishnateja Killamsetty, Harsha Kokel, Rishabh K Iyer

NeurIPS 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical alignment and hypothesis testing · 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
replace Full source training with ORIENT (FLMI) subset (30%)CCSA · Office-31 A->Waccuracy: 0.72 -> 0.76Improved
replace Full source training with ORIENT (FLMI) subset (30%)CCSA · Office-31 D->Waccuracy: 0.93 -> 0.89Hurt
replace Random subset with ORIENT (FLMI) subsetCCSA · Office-31 D->Waccuracy: 0.81 -> 0.89Improved
replace Random subset with ORIENT (FLMI) subsetd-SNE · Office-31 D->Waccuracy: 0.86 -> 0.90Improved
replace Full source training with ORIENT (FLMI) subsetd-SNE · Office-31 D->Aaccuracy: 0.53 -> 0.55Improved
replace Full source training with ORIENT (FLMI) subsetCCSA · Office-Home R->Paccuracy: 0.74 -> 0.78Improved
replace Random subset with ORIENT (FLMI) subsetCCSA · Office-Home R->Caccuracy: 0.47 -> 0.54Improved
replace Full source training with ORIENT (FLMI) subsetCCSA · Office-Home C->Paccuracy: 0.67 -> 0.71Improved
replace Full source training with ORIENT (FLMI) subsetCCSA · Office-Home R->Aaccuracy: 0.57 -> 0.59Improved
replace Full source training with ORIENT (FLMI) subsetd-SNE · Office-Home A->Raccuracy: 0.63 -> 0.61Hurt
replace Full source training with ORIENT (FLMI) subsetd-SNE · Office-Home P->Raccuracy: 0.62 -> 0.63No clear effect
use 30% subset selected by ORIENT instead of full source dataORIENT (FLMI) with d-SNE · Office-31 and Office-Hometraining speedup: ~3×Improved

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