Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Heavy-tailed Representations, Text Polarity Classification & Data Augmentation

Abstract : The dominant approaches to text representation in natural language rely on learning embeddings on massive corpora which have convenient properties such as compositionality and distance preservation. In this paper, we develop a novel method to learn a heavy-tailed embedding with desirable regularity properties regarding the distributional tails, which allows to analyze the points far away from the distribution bulk using the framework of multivariate extreme value theory. In particular, a classifier dedicated to the tails of the proposed embedding is obtained which exhibits a scale invariance property exploited in a novel text generation method for label preserving dataset augmentation. Experiments on synthetic and real text data show the relevance of the proposed framework and confirm that this method generates meaningful sentences with controllable attribute, e.g. positive or negative sentiments.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-02936647
Contributor : Hamid Jalalzai <>
Submitted on : Friday, September 11, 2020 - 3:07:02 PM
Last modification on : Tuesday, May 18, 2021 - 2:08:02 PM

File

Extreme_Representation_for_Nat...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02936647, version 1
  • ARXIV : 2003.11593

Collections

Citation

Hamid Jalalzai, Pierre Colombo, Chloé Clavel, Éric Gaussier, Giovanna Varni, et al.. Heavy-tailed Representations, Text Polarity Classification & Data Augmentation. 2020. ⟨hal-02936647⟩

Share

Metrics

Record views

110

Files downloads

28