Auto-Sizing Neural Networks: With Applications to n-gram Language Models

Article

Abstract

Neural networks have been shown to improve performance across a range of natural-language tasks. However, designing and training them can be complicated. Frequently, researchers resort to repeated experimentation to pick optimal settings. In this paper, we address the issue of choosing the correct number of units in hidden layers. We introduce a method for automatically adjusting network size by pruning out hidden units through L∞,1 and L2,1 regularization. We apply this method to language modeling and demonstrate its ability to correctly choose the number of hidden units while maintaining perplexity. We also include these models in a machine translation decoder and show that these smaller neural models maintain the significant improvements of their unpruned versions.

Attributes

Attribute NameValues
Creator
  • Kenton Murray

  • David Chiang

Publisher
  • Association for Computational Linguistics

Date Created
  • 2015-09-20

Language
  • English

Related Resource(s)
Departments and Units
Record Visibility and Access Public
Content License
  • All rights reserved