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 confidence intervals, hypothesis tests, standard errors, statistical models, likelihood, Bayesian inference, ROC curve, Pearson statistic, goodness of fit tests, and power of tests.
Discusses the Dirichlet distribution, Bayesian inference, posterior mean and variance, conjugate priors, and predictive distribution in the Dirichlet-Multinomial model.
