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.
Table of Contents
Solution Strategy (77x8x8 Representation): The adopted solution, inspired by AlphaZero, is the 77x8x8 representation. ChessMarro has used the same solution within the ChessAIthon competition. This is achieved using 12 separate binary layers (8x8 matrices), one for each piece type and color (eg, one layer only for white pawns, another only for black kings). This ensures that all pieces are treated equally, and the network learns the position and spatial relationship of the pieces, not just their nominal value. CNNs use this structure to learn to recognize complex strategic patterns, such as forks , pins , or checkmate threats .
