These systems are discrete and their state changes are rule based. Example:

Given a chess position, generate a series of moves that will lead to it (there may be many, one, or none, but I only need one valid one)

If there are many examples, (end state, state change sequence) pairs, is there a neural network or other machine learning algorithm that can generate the latter given the former in a way that extrapolates to previously unseen board positions?

In my use cases, it will be possible to find or generate large datasets for both.

  • chess end positions and moves, billions of pairs (e.g. from here)

  • mathematical proofs, 35k+ pairs (see the last line here, the second part is a sequence of tokens/references that proves the first statement)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.