Our debut product, DecodeChess, presents a revolution in computerized chess.
While all chess-playing programs supply the best moves and a mathematical number to indicate the move strength, they do not explain WHY it is the best move.
DecodeChess gives everything that a chess program provides AND explains the chess ideas and the why behind every move.
DecodeChess explains the recommendations of chess-playing programs, which play far better than even the best human players. However, chess-playing programs are based on a “brute-force” search of the tree of possible chess positions. This differs completely from human-evolved conceptual understanding, which enables the player to search for only a few positions from among billions.
Consistent with human comprehension, the DecodeChess algorithm explores the position tree (the phase space) in order to identify the factors responsible for the playing program’s decision. The algorithm extracts hidden causal relations between concepts and constructs explanations in a natural language, alongside graphics, which simulates human comprehension.
Chess is a complex dynamical system wherein any change in the pieces’ position can dramatically change that position’s evaluation, as well as the factors that determine the system’s behavior. Correctly identifying the factors at work is a difficult task requiring keen human cognitive abilities. The DecodeChess algorithm focuses on finding the relevant factors that operate in the specificity of the given concrete position.