Compact Autoregressive Network

Authors

  • Di Wang The University of Hong Kong
  • Feiqing Huang The University of Hong Kong
  • Jingyu Zhao The University of Hong Kong
  • Guodong Li The University of Hong Kong
  • Guangjian Tian Huawei Noah's Ark Lab

DOI:

https://doi.org/10.1609/aaai.v34i04.6079

Abstract

Autoregressive networks can achieve promising performance in many sequence modeling tasks with short-range dependence. However, when handling high-dimensional inputs and outputs, the massive amount of parameters in the network leads to expensive computational cost and low learning efficiency. The problem can be alleviated slightly by introducing one more narrow hidden layer to the network, but the sample size required to achieve a certain training error is still substantial. To address this challenge, we rearrange the weight matrices of a linear autoregressive network into a tensor form, and then make use of Tucker decomposition to represent low-rank structures. This leads to a novel compact autoregressive network, called Tucker AutoRegressive (TAR) net. Interestingly, the TAR net can be applied to sequences with long-range dependence since the dimension along the sequential order is reduced. Theoretical studies show that the TAR net improves the learning efficiency, and requires much fewer samples for model training. Experiments on synthetic and real-world datasets demonstrate the promising performance of the proposed compact network.

Downloads

Published

2020-04-03

How to Cite

Wang, D., Huang, F., Zhao, J., Li, G., & Tian, G. (2020). Compact Autoregressive Network. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 6145-6152. https://doi.org/10.1609/aaai.v34i04.6079

Issue

Section

AAAI Technical Track: Machine Learning