Work

Historical Text Normalization with Delayed Rewards

Year: 2019

Type: conference-paper

Abstract: Training neural sequence-to-sequence models with simple token-level log-likelihood is now a standard approach to historical text normalization, albeit often outperformed by phrasebased models.Policy gradient training enables direct optimization for exact matches, and while the small datasets in historical text normalization are prohibitive of from-scratch reinforcement learning, we show that policy gradient fine-tuning leads to significant improvements across the board.Policy gradient training, in particular, leads to more accurate normalizations for long or unseen words.

Source: Annual Meeting of the Association for Computational Linguistics (ACL)

DOI: 10.18653/v1/p19-1157

Authors: Simon Flachs, Marcel Bollmann, Anders Søgaard

Institution: University of Copenhagen

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: Nvidia, Innovationsfonden, European Commission

Awards: G434337789, Morphologically-informed representations for natural language processing