Poster

Scalarization for Multi-Task and Multi-Domain Learning at Scale

Amelie Royer · Tijmen Blankevoort · Babak Ehteshami Bejnordi

Great Hall & Hall B1+B2 (level 1) #1005
[ ]
Tue 12 Dec 8:45 a.m. PST — 10:45 a.m. PST

Abstract:

Training a single model on multiple input domains and/or output tasks allows for compressing information from multiple sources into a unified backbone hence improves model efficiency. It also enables potential positive knowledge transfer across tasks/domains, leading to improved accuracy and data-efficient training. However, optimizing such networks is a challenge, in particular due to discrepancies between the different tasks or domains: Despite several hypotheses and solutions proposed over the years, recent work has shown that uniform scalarization training, i.e., simply minimizing the average of the task losses, yields on-par performance with more costly SotA optimization methods. This raises the issue of how well we understand the training dynamics of multi-task and multi-domain networks. In this work, we first devise a large-scale unified analysis of multi-domain and multi-task learning to better understand the dynamics of scalarization across varied task/domain combinations and model sizes. Following these insights, we then propose to leverage population-based training to efficiently search for the optimal scalarization weights when dealing with a large number of tasks or domains.

Chat is not available.