Deep Multi-task Representation Learning: A Tensor Factorisation Approach

TL;DR

This paper introduces a deep multi-task learning framework that uses tensor factorisation to automatically learn sharing structures at every layer, improving accuracy and reducing design complexity.

Contribution

It generalises matrix factorisation to tensor factorisation for deep networks, enabling automatic end-to-end knowledge sharing in multi-task learning.

Findings

Higher accuracy in multi-task learning tasks

Fewer design choices needed

Effective for both homogeneous and heterogeneous MTL

Abstract

Most contemporary multi-task learning methods assume linear models. This setting is considered shallow in the era of deep learning. In this paper, we present a new deep multi-task representation learning framework that learns cross-task sharing structure at every layer in a deep network. Our approach is based on generalising the matrix factorisation techniques explicitly or implicitly used by many conventional MTL algorithms to tensor factorisation, to realise automatic learning of end-to-end knowledge sharing in deep networks. This is in contrast to existing deep learning approaches that need a user-defined multi-task sharing strategy. Our approach applies to both homogeneous and heterogeneous MTL. Experiments demonstrate the efficacy of our deep multi-task representation learning in terms of both higher accuracy and fewer design choices.