Deep networks trained on large-scale data can learn transferable features to
promote learning multiple tasks. Since deep features eventually transition from
general to specific along deep networks, a fundamental problem of multi-task
learning is how to exploit the task relatedness underlying parameter tensors
and improve feature transferability in the multiple task-specific layers. This
paper presents Multilinear Relationship Networks (MRN) that discover the task
relationships based on novel tensor normal priors over parameter tensors of
multiple task-specific layers in deep convolutional networks. By jointly
learning transferable features and multilinear relationships of tasks and
features, MRN is able to alleviate the dilemma of negative-transfer in the
feature layers and under-transfer in the classifier layer. Experiments show
that MRN yields state-of-the-art results on three multi-task learning datasets.

本文提出了基于张量正则先验的多线性关系网络 (Multilinear Relationship Networks, MRN)，通过同时学习可转移的特征和任务和特征的多线性关系，发现了深度卷积网络中多个任务特定层参数张量上的任务关系，并解决了特征层中负迁移和分类器层下迁移的困境，实验证明 MRN 在三个多任务学习数据集上的效果优于现有方法。