Hello,
I need to write two programs to train an agent to play a game that is similar to backgammon.
Game rules are little more complicated but still possible.
requirements :
- the implementation on PYTHON of 2 codes the first one using an approach similar to the TD-gammon paper and the second one using a deep learning approach.
- a paper with 10 pages to explain with details of the method and results ... (in a scientific method) WITH DETAILED BIBLIOGRAPHY THAT INCLUDES BOOKS AND PAPERS TO SEE (10 elements )
- a support to understand the codes.