Improving Disentangled Text Representation Learning with Information-Theoretic Guidance

Pengyu Cheng, Martin Renqiang Min, Dinghan Shen, Christopher Malon, Yizhe Zhang, Yitong Li, Lawrence Carin

Research output: Contribution to journalArticlepeer-review

Abstract

Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Similar problems have been studied extensively for other forms of data, such as images and videos. However, the discrete nature of natural language makes the disentangling of textual representations more challenging (e.g., the manipulation over the data space cannot be easily achieved). Inspired by information theory, we propose a novel method that effectively manifests disentangled representations of text, without any supervision on semantics. A new mutual information upper bound is derived and leveraged to measure dependence between style and content. By minimizing this upper bound, the proposed method induces style and content embeddings into two independent low-dimensional spaces. Experiments on both conditional text generation and text-style transfer demonstrate the high quality of our disentangled representation in terms of content and style preservation.
Original languageEnglish (US)
JournalArxiv preprint
StatePublished - Jun 1 2020
Externally publishedYes

Keywords

  • cs.LG
  • stat.ML

Fingerprint

Dive into the research topics of 'Improving Disentangled Text Representation Learning with Information-Theoretic Guidance'. Together they form a unique fingerprint.

Cite this