I have trouble seeing reinforcement learning as convolution, for example.
Reply on news.ycombinator.com