Deep Multi-task Representation Learning: A Tensor Factorisation Approach

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.

9 pages, Accepted to ICLR 2017 Conference Track. This is a conference version of the paper. For the multi-domain learning part (not in this version), please refer to https://arxiv.org/pdf/1605.06391v1.pdf

Cited by in corpus (41)

• An Overview of Multi-Task Learning in Deep Neural Networks
• A Survey on Multi-Task Learning
• Tensor Networks for Dimensionality Reduction and Large-Scale Optimizations. Part 2 Applications and Future Perspectives
• Tensor Methods in Computer Vision and Deep Learning
• Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference
• MEDIC: A Multi-Task Learning Dataset for Disaster Image Classification
• Convolutional Tensor-Train LSTM for Spatio-temporal Learning
• Tensor Regression Networks
• Trace Norm Regularised Deep Multi-Task Learning
• Detecting 32 Pedestrian Attributes for Autonomous Vehicles
• Multi-Task Zipping via Layer-wise Neuron Sharing
• Pseudo-task Augmentation: From Deep Multitask Learning to Intratask Sharing---and Back
• Controllable Pareto Multi-Task Learning
• Beyond Shared Hierarchies: Deep Multitask Learning through Soft Layer Ordering
• Tensor Contraction Layers for Parsimonious Deep Nets
• MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning
• Experimental Results of Underwater Sound Speed Profile Inversion by Few-shot Multi-task Learning
• Evolutionary Architecture Search For Deep Multitask Networks
• A Simple General Approach to Balance Task Difficulty in Multi-Task Learning
• Meta-Learning Multi-task Communication
• Multi-task neural networks by learned contextual inputs
• Modeling the Sequential Dependence among Audience Multi-step Conversions with Multi-task Learning in Targeted Display Advertising
• Auxiliary Learning for Deep Multi-task Learning
• DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics
• Tensor Decompositions in Deep Learning
• Modular Universal Reparameterization: Deep Multi-task Learning Across Diverse Domains
• Deeper, Broader and Artier Domain Generalization
• Boosting Multitask Learning on Graphs through Higher-Order Task Affinities
• Deep Multi-Task Learning via Generalized Tensor Trace Norm
• Marginalizable Density Models
• Tensor-Train Networks for Learning Predictive Modeling of Multidimensional Data
• Provable Adaptation across Multiway Domains via Representation Learning
• Convolutional Neural Networks with Transformed Input based on Robust Tensor Network Decomposition
• Compositional Models: Multi-Task Learning and Knowledge Transfer with Modular Networks
• Adaptive Activation Network and Functional Regularization for Efficient and Flexible Deep Multi-Task Learning
• Deep Asymmetric Multi-task Feature Learning
• Deep Multi-Task Augmented Feature Learning via Hierarchical Graph Neural Network
• Multi-Task Generative Adversarial Nets with Shared Memory for Cross-Domain Coordination Control
• Multiple Classification with Split Learning
• From Persistent Homology to Reinforcement Learning with Applications for Retail Banking
• Deep Stacking Networks for Low-Resource Chinese Word Segmentation with Transfer Learning