Debated backpropagation

Isaac James*, Christopher Luciano Stone*, Alice Toniolo*

*Corresponding author for this work

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

19 Downloads (Pure)

Abstract

Dialogue has long been used in human society to explain seemingly opaque concepts. In this paper we focus on how to better explain training models for neural networks, to entertain as well as inform. We present a multi-agent argumentation-based dialogue system to generate human understandable dialogue to explain backpropagation. The system incorporates a model of agent personality and introduces social elements between agents to produce characterful discussion. Natural language templates are used to render utterances in English.
Original languageEnglish
Title of host publicationArtificial Intelligence and Cognitive Science 2021
Subtitle of host publicationThe 29th Irish Conference on Artificial Intelligence and Cognitive Science 2021 Dublin, Republic of Ireland, December 9-10, 2021
EditorsArjun Pakrashi, Ellen Rushe, Mehran Hossein Zadeh Bazargani, Brian Mac Namee
Pages260-271
Number of pages12
Publication statusPublished - 11 Mar 2022
Event29th Irish Conference on Artificial Intelligence and Cognitive Science - University College Dublin, Dublin, Ireland
Duration: 9 Dec 201910 Dec 2021
Conference number: 29
https://aics2021.ucd.ie

Publication series

NameCEUR Workshop Proceedings
ISSN (Electronic)1613-0073

Conference

Conference29th Irish Conference on Artificial Intelligence and Cognitive Science
Abbreviated titleAICS21
Country/TerritoryIreland
CityDublin
Period9/12/1910/12/21
Internet address

Keywords

  • Backpropogation
  • Explanation
  • Argumentation
  • Dialogue
  • NLG

Fingerprint

Dive into the research topics of 'Debated backpropagation'. Together they form a unique fingerprint.

Cite this