27-28 November, 2023
The purpose of the workshop Analogies: From learning to explainability is to bring together AI researchers at the cross roads of machine learning, natural language processing, knowledge presentation and reasoning, who are interested in the various applications of analogical reasoning in machine learning, especially, for explainability purposes.
This workshop is supported by the working group MyHIA and Explicon of the GDR RADIA, and by the ANR project AT2TA (2023-2026).
Organisers: Miguel Couceiro (miguel.couceiro@loria.fr) & Zied Bouraoui (zied.bouraoui@cril.fr)
Date and location: The workshop will take place on 27-28 November 2023 at the Université d’Artois in Arras and in hybrid mode. Building: Maison de la recherche in Arras.
Attendance: The event is free to attend, but registration is mandatory by November 16 using this link.
Submissions: Just send to the organisers your title and abstract by November 16, 2023.
Program:
Monday (27/11/2023): Link
13:30-14h00: Welcome and presentation (Miguel & Zied)
14:00-15:00: Lonneke van der Plas (Idiap, Martigny, Switzerland)
Can language models learn analogical reasoning? Investigating training objectives and comparisons to human performance.
15:00-16:00: Fadi Badra (LIMICS, Université Sorbonne Paris Nord)
Learning by Analogy with the CoAT Method: Modeling and Perspectives
16:00-16:30: Break
16:30 – 17:30: Henri Prade (IRIT, Toulouse)
Explanations by analogical proportions and analogical relevance index
17:30 -18:30: Pierre Monnin (Wimmics, Université Côte d’Azur)
Knowledge Graphs and Analogical Reasoning: from the Example of Zero-Shot KG Bootstrapping to Perspectives
Tuesday (28/11/2023): Link
9:00-10:00: Jean Lieber (Loria, Nancy)
About links between Case-Based Reasoning and Analogy
10:00 -10:30: break
10:30-11:30: Emiliano Lorini (IRIT, Toulouse)
Non-Classical Logics for Explanations in AI Systems
11:30-12:30: Yves Lepage (Université Waseda, Japan)
Analogie, explication des données de langue et travaux récents sur représentations vectorielles de phrases et analogie
12:30-14:00: Lunch
14:00-15:00: Myriam Bounhas (Liwa College of Technology, Abu Dhabi)
Analogical Classification using Competent Data Pairs
15:00- 18:00: Discussions
