English

GO4Align: Group Optimization for Multi-Task Alignment

Machine Learning 2024-10-30 v2 Computer Vision and Pattern Recognition

Abstract

This paper proposes \textit{GO4Align}, a multi-task optimization approach that tackles task imbalance by explicitly aligning the optimization across tasks. To achieve this, we design an adaptive group risk minimization strategy, comprising two techniques in implementation: (i) dynamical group assignment, which clusters similar tasks based on task interactions; (ii) risk-guided group indicators, which exploit consistent task correlations with risk information from previous iterations. Comprehensive experimental results on diverse benchmarks demonstrate our method's performance superiority with even lower computational costs.

Keywords

Cite

@article{arxiv.2404.06486,
  title  = {GO4Align: Group Optimization for Multi-Task Alignment},
  author = {Jiayi Shen and Cheems Wang and Zehao Xiao and Nanne Van Noord and Marcel Worring},
  journal= {arXiv preprint arXiv:2404.06486},
  year   = {2024}
}
R2 v1 2026-06-28T15:49:05.949Z