In the history of infectious diseases, a single drug has been easily resisted by mutations in the pathogen. In contrast, when multiple drugs are used simultaneously, each helps prevent the emergence of resistance to the others. Ryfamate derives from Rifamate which is the combination of two medicines for tuberculosis to retard the development of drug resistance. As with this method, Ryfamate aims to combine an NNUE alpha-beta search engine with deep learning MCTS outcomes to compensate for each other's shortcomings.
Recently, with the advent of deep learning, the opening of games has become more important. Ryfamate selects book moves with the highest hit rates based on the number of playouts instead of those with the best values. This selection not only improves the confidence of the move but also saves thinking time and guides the game to a specific pattern. As a result, Ryfamate strives to gain an advantage even against players with more computational resources.
YaneuraOu : https://github.com/yaneurao/YaneuraOu
dlshogi : https://github.com/TadaoYamaoka/DeepLearningShogi