- #1
Bob3141592
- 236
- 2
Does anybody in this forum play chess? Or have an idea of how computer chess programs work?
Even the very best of them, and lately they have gotten very, very good, work more or less the same way, and always have. In those algorithms, after the move generators and transposition tables and search pruning, are evaluation functions that I find arbitrary, and based on human judgement applied in domians without real theoretical justification.
I have a different idea, but no clue about how to actually go about it. I wonder if it wouldn't be possible to model each chess piece as some kind of a quantum particle. The rules of the game would be entirely contained in these models (such as en passant, castling restrictions, etc).
Then, in order to determine the best move, perhaps it wouldn't be necessary to look ahead at all. Perhaps it would only be necessary to allow the wave states of the pieces to interact and solve for the maximum overall energy for the right move to present itself.
How would one go about creating a quantum wave function for the behavior of a chess piece? Is it possible?
This idea might not work at all. Probably wouldn't. But if it does work, it might work very well. At least it'd be different, qualitatively different. And, it might be fun.
And ideas?
Even the very best of them, and lately they have gotten very, very good, work more or less the same way, and always have. In those algorithms, after the move generators and transposition tables and search pruning, are evaluation functions that I find arbitrary, and based on human judgement applied in domians without real theoretical justification.
I have a different idea, but no clue about how to actually go about it. I wonder if it wouldn't be possible to model each chess piece as some kind of a quantum particle. The rules of the game would be entirely contained in these models (such as en passant, castling restrictions, etc).
Then, in order to determine the best move, perhaps it wouldn't be necessary to look ahead at all. Perhaps it would only be necessary to allow the wave states of the pieces to interact and solve for the maximum overall energy for the right move to present itself.
How would one go about creating a quantum wave function for the behavior of a chess piece? Is it possible?
This idea might not work at all. Probably wouldn't. But if it does work, it might work very well. At least it'd be different, qualitatively different. And, it might be fun.
And ideas?