The ChessAIThon project (2025-1-ES01-KA220-VET-000354329) is co-funded by the European Union. The views and opinions expressed in this publication are those of the author(s) only and do not necessarily reflect those of the European Union or the Spanish Service for the Internationalisation of Education (SEPIE). Neither the European Union nor the National Agency SEPIE can be held responsible for them.

Deep learning and data analysis concepts for chess movement solving

Home

Table of Contents

Part 2

Practical Example of Error: For the CNN to function, the board representation must be numerical. A simple representation using a single matrix where pieces are encoded with different numerical values (eg, 6 for the King, 1 for the Pawn) should be avoided, as the CNN might interpret that higher numbers mean more important pieces, which is not correct in all strategic situations.