Hi, I am working on a program to play a game which is structurally similar to backgammon. (It is a proprietary game, so I don't want to use the name on a semi-public forum.) I anticipate using some form of machine learning to get the program to play well. All of the references which I have been able to find on machine learning are for training against static data sets or solitaire games. I have some ideas on how to do what I want, but not necessarily using state-of-the-art techniques. Because I am hoping, in addition to coming up with a good program, to be in a position to get a job in the machine learning field, I am looking for any resources on current techniques that might be helpful in this domain. It would be particularly helpful to use AWS machine learning here, as I am also trying to get up to speed on AWS. --Alan