Controllable Pareto Multi-Task Learning

Xi Lin, Zhiyuan Yang, Qingfu Zhang, Sam Kwong

Keywords: Multi-Task Learning, Multi-Objective Optimization

Abstract: A multi-task learning (MTL) system aims at solving multiple related tasks at the same time. With a fixed model capacity, the tasks would be conflicted with each other, and the system usually has to make a trade-off among learning all of them together. Multiple models with different preferences over tasks have to be trained and stored for many real-world applications where the trade-off has to be made online. This work proposes a novel controllable Pareto multi-task learning framework, to enable the system to make real-time trade-off switch among different tasks with a single model. To be specific, we formulate the MTL as a preference-conditioned multiobjective optimization problem, for which there is a parametric mapping from the preferences to the Pareto stationary solutions. A single hypernetwork-based multi-task neural network is built to learn all tasks with different trade-off preferences among them, where the hypernetwork generates the model parameters conditioned on the preference. At the inference time, MTL practitioners can easily control the model performance based on different trade-off preferences in real-time. Experiments on different applications demonstrate that the proposed model is efficient for solving various multi-task learning problems.

One-sentence Summary: This work proposes a novel approach to learn the entire trade-off curve for MTL problems.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:2010.06313/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=7cjBiwRdG4