![[2602.18591] Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.18591] Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning
Summary
The paper presents ETAP, a framework for predicting task affinity in multi-task learning, enhancing efficiency by grouping tasks that benefit from joint learning.
Why It Matters
Efficient multi-task learning is crucial for optimizing model performance in machine learning. This research addresses the challenge of identifying beneficial task groupings, which can significantly enhance learning outcomes and resource utilization in various applications.
Key Takeaways
- ETAP predicts task affinity to improve multi-task learning efficiency.
- The framework combines gradient-based and data-driven estimators for accurate predictions.
- ETAP outperforms existing methods in predicting performance gains across diverse tasks.
- Understanding task relationships can lead to better model training strategies.
- The approach is scalable and applicable to large sets of tasks.
Computer Science > Machine Learning arXiv:2602.18591 (cs) [Submitted on 20 Feb 2026] Title:Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning Authors:Afiya Ayman, Ayan Mukhopadhyay, Aron Laszka View a PDF of the paper titled Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning, by Afiya Ayman and 2 other authors View PDF Abstract:A fundamental problem in multi-task learning (MTL) is identifying groups of tasks that should be learned together. Since training MTL models for all possible combinations of tasks is prohibitively expensive for large task sets, a crucial component of efficient and effective task grouping is predicting whether a group of tasks would benefit from learning together, measured as per-task performance gain over single-task learning. In this paper, we propose ETAP (Ensemble Task Affinity Predictor), a scalable framework that integrates principled and data-driven estimators to predict MTL performance gains. First, we consider the gradient-based updates of shared parameters in an MTL model to measure the affinity between a pair of tasks as the similarity between the parameter updates based on these tasks. This linear estimator, which we call affinity score, naturally extends to estimating affinity within a group of tasks. Second, to refine these estimates, we train predictors that apply non-linear transformations and correct residual errors, capturing complex and non-linear task relationships. We train these predictors...