Statistics, Probability and Machine Learning Short Course

Here is a Statistics, Probability and Machine Learning course I developed mainly for PhD students at UTS. It's constantly being updated. I enjoy sharing my knowledge with other researchers and industry practitioners. Depends on the level of the participants, I am happy to give one to ten days face-to-face tutorials / workshops on these topics. Please email me. This short course is focused mainly on probabilistic generative models. Machine Learning of course, contains much more topics than what these notes cover.

大家好，從2015年10月開始，我為國(guó)內(nèi)的同行，同學(xué)們做了一系列概率機(jī)器學(xué)習(xí)的視頻公開課,。我每星期都會(huì)更新。為了方便大家，我是用中文講的. The English version of my machine learning MOOCS is also on its way...

	優(yōu)酷 連接
	Youtube 連接

	Intro to Bayesian Statistics     General probability knowledge, Bayesian Statistics
	Probabilities and Estimations     Various common probability distribution (functions), natural parameters and Maximum Likelihood & Maximum A Posterior Estimation
	Statistical Properties     Various useful statistical properties include inequalities, convergence and uniqueness
	Expectation-Maximization     Proof of convergence for E-M, examples of E-M through Gaussian Mixture Model
	Continous and Discrete State Dynamic Systems     Derivations for Kalman Filter and Hidden Markov Model
	Monte Carlo and Sequential Monte Carlo Inference     Overview of several Sampling techniques, including Rejection, Adaptive Rejection, Importance, Markov Chain Monte Carlo (MCMC), Gibbs, Bootstrap Particle Filter, and Auxiliary Particle Filter
	Non-parametric Bayes & applications     Dirichlet Process, Hieratical Dirichlet Process, HDP-HMM, Indian Buffet Process, and applications of DP to relational models.
	Variational Bayes     Explain Variational Bayes in two parts: non-exponential and exponential family distribution plus stochastic variational inference.
	Linear Regression and Support Vector Machine     Explain Linear Model, Norms and the internal workings of support vector machine.

 

Some older notes (before 2009) for Computer vision

Camera calibration

Mean-shift Tracking  (2 hours face to face)

Non-Linear Least Square Fitting (Part A)