Computer-Generated Chess Problem 03085
An original 'KQRBNNPP vs kqr' chess problem generated autonomously by a computer program, Chesthetica, using the Digital Synaptic Neural Substrate (DSNS) computational creativity approach. The DSNS does not use endgame tablebases, neural networks or any kind of machine learning found in traditional artificial intelligence (AI). It also has nothing to do with deep learning. There is no known limit to the quantity or type of compositions that can be generated. The largest (Lomonosov) tablebase today is for 7 pieces which contains over 500 trillion positions. With each additional piece, the number of possible positions increases exponentially. It is therefore impossible that this problem with 11 pieces could have been taken from such a database.
White to Play and Mate in 4
Chesthetica v11.98 (Selangor, Malaysia)
Generated on 16 Oct 2020 at 3:19:58 AM
Humans have been composing original chess problems for over a thousand years. Now a computer can do it too. White has a decisive material advantage in this position but the winning sequence may not be immediately clear. Leave a comment below, if you like. Collectively, these puzzles are intended to cater to players of all levels.
Solution