The lecture slides will be released before the class.
[Home] | [Information] | [Topics] | [Lectures] | [Projects] |
The lecture slides will be released before the class.
Date | Topic | Reading Materials | Assignment |
---|---|---|---|
Tue 1.09 | Class Overview: Probabilistic Machine Leanring Definition, Course Topics | HW0 Released, Due 01/19 | |
Thu 1.11 | Class Overview: Course Policy, Basics Review | ||
Tue 1.16 | Basics Overview: Matrix/Vector Derivatives, Convex Functions | Probability: Probability Space, Independenc | ||
Thu 1.19 | Basics Overview:Convex Conjugate | Probability: Probability Space, Independenc | PRML Ch10.5 pp.493-496 | |
Tue 1.22 | Probability: CDF, PDF, Joint Dist., Condition Dist | Probability distributions: MLE, MAP, Bernoulli, Binomial, Categorical | ||
Thu 1.25 | Probability distributions: Beta, Dirichlet, Multivariate Gaussian | PRML Ch2.1-2.3 | |
Tue 1.30 | Probability distributions: Gamma, Wishart, Student t, Multivariate Student t | Conjugate Prior: Beta | PRML Ch1.6, 2.4 | HW1 Released, Due on 02/16 |
Tue 2.1 | Conjugate prior: Gamma, Dirichlet, Wishart, exponential family | Information Theory: Entropy | PRML Ch1.6, 2.4, 2.4.3, Jordan's slides[1][2] | |
Tue 2.6 | Information Theory:Differential Entropy, KL Diveregence | Noninformative priors: Bayesian Philosophy, Uniform, Translation Invaraince | PRML Ch1.6, 2.4, 2.4.3, Jordan's slides[1][2] | Mid-term Project Report Due 03/15 |
Thu 2.9 | Noninformative priors: Jefferys Prior, Exchangeability, De Finetti’s Theorem | PRML Ch3 | |
Tue 2.13 | Noninformative priors: De Finetti’s Theorem | Generalized Linear Models: Design matrix, MLE, Regularization, Bayesian Linear Regression | PRML Ch3 | |
Thu 2.15 | Generalized Linear Models: posterior of weights, predictive distribution, evidence maximization, type II MLE | PRML Ch3 | |
Tue 2.20 | Generalized Linear Models: logistic regression, multi-class, probit regression, ordinal regression, generalized linear models | PRML Ch3, 4 | HW2 Released, Due on 03/15 |
Thu 2.22 | Probabilistic Graphical Models: Bayesian Networks, Conditional Indepdence, D-separation | PRML Ch8 | |
Tue 2.27 | Probabilistic Graphical Models: D-separation, Bayes ball algorithm, Markov Blanket, Markov Random Fields, Moralization | Inference: Tasks | PRML Ch8 | |
Thu 2.29 | Inference: Forward and Backward messages, Factor Graphs, Sum-Product Algorithm | PRML Ch8 | |
Tue 3.05 | |||
Thu 3.07 | |||
Tue 3.12 | Inference: Sum-Product Algorithm, Implementation, Example | PRML Ch8 | |
Thu 3.14 | Inference: Max-Sum Algorithm | Approximate Inference | Laplace's Approximation: General Ideas, Bayesian Logistic Regression | PRML Ch8 | |
Tue 3.19 | Laplace's Approximation: Bayesian Logistic Regression | Variational Inference: Gaussian Mixture Model, EM Algorithm | PRML Ch8 | HW3 Released, Due 03/29 |
Thu 3.21 | |||
Tue 3.26 | Variational Inference: Global Variational Inference, Local Variational Inference, Bayesian Logistic Regression | PRML Ch10.5, 10.6 | |
Thu 3.28 | Variational Inference: Bayesian Logistic Regression, Variational Message Passing | PRML Ch10.4 Blei's paper | |
Tue 4.02 | |||
Thu 4.04 | Latent Dirichlet Allocation|Markov Chain Monte Carlo Sampling: Markov Chains, Basic Framework | PRML Ch11.2 | |
Tue 4.09 | Markov Chain Monte Carlo Sampling: Invariant/Stationary Dist., Transition Kernel, Ergodicity, Detailed Balance, Metropolis-Hastings, Gibbs Sampling | PRML Ch11.3 | HW4 Released, Due on 04/18 |
Thu 4.11 | Markov Chain Monte Carlo Sampling: Hamiltonian System, Reversability, Conservation, Volume Preservation, Theory of using dynamics | Neal's intro, Youhan Fang's thesis Ch2 | |
Tue 4.16 | Markov Chain Monte Carlo Sampling: Leap Frog, Wrap-up | Bayesian Neural Networks: Neural Networks, Forward Pass, Stochastic Optimization, Bayes by BP, Reparameterization Trick |