Here is a new 'KRRNN vs kqrpp' four-move chess problem generated by the prototype computer program, Chesthetica, using the 'Digital Synaptic Neural Substrate' computational creativity method. It does not use endgame tablebases, artificial neural networks, machine learning or any kind of typical AI. Depending on the type and complexity of the problem desired, a single instance of Chesthetica running on a desktop computer can probably generate anywhere between one and ten problems per hour.
White to Play and Mate in 4
Chesthetica v10.67 : Selangor, Malaysia
2018.5.27 8:00:07 AM
Material is even. Try to solve this as quickly as you can. If you like it, please share with your friends. Take some time to study the analysis and you might appreciate the puzzle a little more.
Solution (Skip to 0:35)
< | Facebook | Twitter | Kindle E-book | Website | Reddit| >