Multi-Task Learning as Multi-Objective Optimization

TL;DR

This paper reformulates multi-task learning as a multi-objective optimization problem, proposing an efficient method to find Pareto optimal solutions that outperform existing approaches in various deep learning tasks.

Contribution

It introduces a scalable approach to multi-objective optimization in multi-task learning, enabling Pareto optimal solutions through an upper bound optimization technique.

Findings

Achieves higher performance than recent multi-task learning methods.

Effectively handles conflicting tasks in deep learning applications.

Provides theoretical guarantees for Pareto optimality.

Abstract

In multi-task learning, multiple tasks are solved jointly, sharing inductive bias between them. Multi-task learning is inherently a multi-objective problem because different tasks may conflict, necessitating a trade-off. A common compromise is to optimize a proxy objective that minimizes a weighted linear combination of per-task losses. However, this workaround is only valid when the tasks do not compete, which is rarely the case. In this paper, we explicitly cast multi-task learning as multi-objective optimization, with the overall objective of finding a Pareto optimal solution. To this end, we use algorithms developed in the gradient-based multi-objective optimization literature. These algorithms are not directly applicable to large-scale learning problems since they scale poorly with the dimensionality of the gradients and the number of tasks. We therefore propose an upper bound for the multi-objective loss and show that it can be optimized efficiently. We further prove that optimizing this upper bound yields a Pareto optimal solution under realistic assumptions. We apply our method to a variety of multi-task deep learning problems including digit classification, scene understanding (joint semantic segmentation, instance segmentation, and depth estimation), and multi-label classification. Our method produces higher-performing models than recent multi-task learning formulations or per-task training.