Delves into the fundamental limits of gradient-based learning on neural networks, covering topics such as binomial theorem, exponential series, and moment-generating functions.
Covers Markov processes, transition densities, and distribution conditional on information, discussing classification of states and stationary distributions.
