arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.15638cs.LG
收藏

多任务学习中的公平资源分配

Fair Resource Allocation in Multi-Task Learning

University at Buffalo(布法罗大学)

Hao Ban, Kaiyi Ji

AI总结:

本文借鉴通信网络公平资源分配,将多任务学习建模为效用最大化问题,提出具备收敛保证的 FairGrad,并在监督与强化学习基准上达到最先进性能。

摘要:

通过联合学习多个任务,多任务学习(MTL)能够利用跨任务的共享知识,从而提升数据效率和泛化性能。然而,MTL 的一个主要挑战在于存在冲突梯度,这会妨碍某些任务得到公平优化,进而限制 MTL 取得更好整体性能的能力。受通信网络中公平资源分配的启发,我们将 MTL 的优化表述为一个效用最大化问题,在不同公平性度量下最大化各任务的损失下降量。为解决该问题,我们提出 FairGrad,这是一种新颖的 MTL 优化方法。FairGrad 不仅能够灵活强调某些任务,还实现了理论收敛保证。大量实验表明,在监督学习和强化学习的一系列多任务基准上,我们的方法在梯度操纵方法中能够达到最先进性能。此外,我们将 α-fairness(α-公平性)思想融入多种 MTL 方法的损失函数中。广泛的实证研究表明,这些方法的性能可以得到显著增强。代码位于 https://github.com/OptMN-Lab/fairgrad。

英文摘要:

By jointly learning multiple tasks, multi-task learning (MTL) can leverage the shared knowledge across tasks, resulting in improved data efficiency and generalization performance. However, a major challenge in MTL lies in the presence of conflicting gradients, which can hinder the fair optimization of some tasks and subsequently impede MTL's ability to achieve better overall performance. Inspired by fair resource allocation in communication networks, we formulate the optimization of MTL as a utility maximization problem, where the loss decreases across tasks are maximized under different fairness measurements. To solve this problem, we propose FairGrad, a novel MTL optimization method. FairGrad not only enables flexible emphasis on certain tasks but also achieves a theoretical convergence guarantee. Extensive experiments demonstrate that our method can achieve state-of-the-art performance among gradient manipulation methods on a suite of multi-task benchmarks in supervised learning and reinforcement learning. Furthermore, we incorporate the idea of $α$-fairness into loss functions of various MTL methods. Extensive empirical studies demonstrate that their performance can be significantly enhanced. Code is provided at \url{https://github.com/OptMN-Lab/fairgrad}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑