Towards the use of deep reinforcement learning with global policy For query-based extractive summarisation

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

Supervised approaches for text summarisation suffer from the problem of mismatch between the target labels/scores of individual sentences and the evaluation score of the final summary. Reinforcement learning can solve this problem by providing a learning mechanism that uses the score of the final summary as a guide to determine the decisions made at the time of selection of each sentence. In this paper we present a proof-of-concept approach that applies a policy-gradient algorithm to learn a stochastic policy using an undiscounted reward. The method has been applied to a policy consisting of a simple neural network and simple features. The resulting deep reinforcement learning system is able to learn a global policy and obtain encouraging results.
Original languageEnglish
Title of host publicationAustralasian Language Technology Association Workshop 2017
Subtitle of host publicationProceedings of the Workshop
EditorsJojo Sze-Meng Wong, Gholamreza Haffari
Pages103-107
Number of pages5
Publication statusPublished - 2017
EventAustralasian Language Technology Association Workshop 2017 - Brisbane, Australia
Duration: 6 Dec 20178 Dec 2017

Conference

ConferenceAustralasian Language Technology Association Workshop 2017
CountryAustralia
CityBrisbane
Period6/12/178/12/17

Fingerprint Dive into the research topics of 'Towards the use of deep reinforcement learning with global policy For query-based extractive summarisation'. Together they form a unique fingerprint.

Cite this