So, use a random strategy and measure the entropy of your opponent's behavior. And modify accordingly, as needed. A lot of the best ml right now is effectively

So, use a random strategy and measure the entropy of your opponent's behavior. And modify accordingly, as needed. A lot of the best ml right now is effectively about making better conditional probability distributions. You always get random output, but skewed according to the circumstances, and sharp according to confidence in the result.