AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Deconstructing Multi-Task Active Learning: The Paradox of Gradient Conflict and Orthogonal Decomposition

Juho Jung, Seongjae Kang, Jinyoung Han

NeurIPS 2026 · Learning theory and algorithms › Combinatorial and search optimization · approach: Multi-objective optimization · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Argues that linear scalarization of heterogeneous active-learning objectives (uncertainty, diversity, gradient conflict) in multi-task active learning is fundamentally limited and proposes moving from combination to an orthogonal decomposition view.

Problem

Multi-task active learning (MTL-AL) methods typically combine heterogeneous sample-selection objectives such as uncertainty, diversity, and gradient conflict via linear scalarization, which may fail to capture Pareto-optimal trade-offs under conflicting objectives.

Approach

Replaces the entangled linear combination of selection objectives with an orthogonal decomposition formulation, shifting from combining objectives to decomposing them so that conflicting signals can be handled Pareto-optimally.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
penalize gradient conflict in the acquisition score vs ignoring it (ablation)MTL active learning (AdamW) · COCOMTL composite: 0.0756 -> 0.0589Hurt
embrace conflict-informed selection with PCGrad vs ignoring conflict (ablation)ODS without Stage 2 · COCOMTL composite: 0.0756 -> 0.0932Improved
add Stage 2 k-center coverage after informativeness filtering (ablation)ODS · COCOMTL composite: 0.0932 -> 0.105Improved
add CTMI to Hardness in the informativeness score (ablation)ODS · COCOMTL composite: 0.094 -> 0.105Improved
change combination rule of Hardness and CTMI (rank-sum, z-score, product) (ablation)ODS · COCOMTL composite: 0.723-0.731No clear effect
two-stage orthogonal decomposition vs linear scalarization of the same scores (ablation)ODS · COCO seed 42MTL composite: +6.5%Improved
use PCGrad instead of AdamW with ODS (ablation)ODS · COCOdelta m (%): +8.2 -> +10.5Improved
ODS versus MT-BADGE under the same optimizerODS · COCOdelta m (%): +7.5 -> +10.5 (PCGrad)Improved
entropy or CoreSet selection vs randomEntropy, CoreSet · COCOMTL composite: -0.005, -0.002Hurt
ODS versus adapted two-stage methods (FASS, Cluster-Margin, PPAL, D2-Pruning, SES)ODS · COCO unified protocolbest delta m (%): 4.41 vs 2.92-4.15No clear effect
ODS versus adapted two-stage methods on whole trajectoryODS · COCO unified protocoltrajectory delta m (%): +1.03 vs -0.15 to +0.41Improved
ODS versus random sampling on other datasetsODS · Cityscapes, NYUv2delta m (%): +8.8, +7.5Improved

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