Designing Logic Tensor Networks for Visual Sudoku puzzle classification

Published in 17th International Workshop on Neural-Symbolic Learning and Reasoning (NeSy 2023), 2023

Recommended citation: Designing Logic Tensor Networks for Visual Sudoku puzzle classification / Morra, Lia; Azzari, Alberto; Bergamasco, Letizia; Braga, Marco; Capogrosso, Luigi; Delrio, Federico; DI GIACOMO, Giuseppe; Eiraudo, Simone; Ghione, Giorgia; Giudice, Rocco; Koudounas, Alkis; Piano, Luca; REGE CAMBRIN, Daniele; Risso, Matteo; Rondina, Marco; Russo, ALESSANDRO SEBASTIAN; Russo, Marco; Taioli, Francesco; Vaiani, Lorenzo; Vercellino, Chiara. - ELETTRONICO. - 3432:(2023), pp. 223-232. (Intervento presentato al convegno 17th International Workshop on Neural-Symbolic Learning and Reasoning (NeSy 2023) tenutosi a Certosa di Pontignano, Siena (Italia) nel July 3-5, 2023). https://hdl.handle.net/11583/2978475

Given the increasing importance of the neurosymbolic (NeSy) approach in artificial intelligence, there is a growing interest in studying benchmarks specifically designed to emphasize the ability of AI systems to combine low-level representation learning with high-level symbolic reasoning. One such recent benchmark is Visual Sudoku Puzzle Classification, that combines visual perception with relational constraints. In this work, we investigate the application of Logic Tensork Networks (LTNs) to the Visual Sudoku Classification task and discuss various alternatives in terms of logical constraint formulation, integration with the perceptual module and training procedure.