arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2206.08403cs.LGcs.AIcs.CY
收藏

学习教授公平意识深度多任务学习

Learning to Teach Fairness-aware Deep Multi-task Learning

L3S Research Center(L3S研究中心);Leibniz University Hannover(汉诺威莱布尼茨大学);Free University Berlin(柏林自由大学)

Arjun Roy, Eirini Ntoutsi

摘要:

公平意识学习主要关注单任务学习(STL)。多任务学习(MTL)的公平性影响最近才被考虑,并提出了一种开创性的方法,该方法考虑了每个任务的公平性-准确性权衡以及不同任务之间的性能权衡。我们提出了一种灵活的方法,而不是一个刚性的公平性和准确性的权衡制定,学习如何在MTL设置公平,选择目标(准确性或公平性),以优化在每一步。我们介绍了L2 T-FMT算法,这是一个教师-学生网络协同训练;学生学习解决公平MTL问题,而教师指导学生从准确性或公平性中学习,这取决于每个任务的学习难度。此外,这种在每个任务的每个步骤使用哪个目标的动态选择将权衡权重的数量从2 T减少到T,其中T是任务的数量。我们在三个真实数据集上的实验表明,L2 T-FMT在公平性(12-19%)和准确性(高达2%)方面都优于最先进的方法。

英文摘要:

Fairness-aware learning mainly focuses on single task learning (STL). The fairness implications of multi-task learning (MTL) have only recently been considered and a seminal approach has been proposed that considers the fairness-accuracy trade-off for each task and the performance trade-off among different tasks. Instead of a rigid fairness-accuracy trade-off formulation, we propose a flexible approach that learns how to be fair in a MTL setting by selecting which objective (accuracy or fairness) to optimize at each step. We introduce the L2T-FMT algorithm that is a teacher-student network trained collaboratively; the student learns to solve the fair MTL problem while the teacher instructs the student to learn from either accuracy or fairness, depending on what is harder to learn for each task. Moreover, this dynamic selection of which objective to use at each step for each task reduces the number of trade-off weights from 2T to T, where T is the number of tasks. Our experiments on three real datasets show that L2T-FMT improves on both fairness (12-19%) and accuracy (up to 2%) over state-of-the-art approaches.

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。

↑