Abstract: Tree-structured multi-task architectures have been employed to jointly tackle multiple vision tasks in the context of multi-task learning (MTL). The major challenge is to determine where to branch out for each task given a backbone model to optimize for both task accuracy and computation efficiency. To address the challenge, this paper proposes a recommender that, given a set of tasks and a convolutional neural network-based backbone model, automatically suggests tree-structured multi-task architectures that could achieve a high task performance while meeting a user-specified computation budget without performing model training. Extensive evaluations on popular MTL benchmarks show that the recommended architectures could achieve competitive task accuracy and computation efficiency compared with state-of-the-art MTL methods. Our tree-structured multi-task model recommender is open-sourced and available at https://github.com/zhanglijun95/TreeMTL.
Keywords: Multi-Task Learning, Architecture Recommendation Framework
One-sentence Summary: A recommender for automatically suggesting tree-structured multi-task architectures with high task performance while meeting the computational budget.
Track: Main track
Reproducibility Checklist: Yes
Broader Impact Statement: Yes
Paper Availability And License: Yes
Code Of Conduct: Yes
Reviewers: Lijun Zhang, firstname.lastname@example.org Xiao Liu, email@example.com
Main Paper And Supplementary Material: pdf
CPU Hours: 0
GPU Hours: 0
TPU Hours: 0