Labs and Homeworks for Machine Learning course, MSc AI @ UvA 2018/2019.

Solutions and implementation from Davide Belli and Gabriele Cesa.

Lab 1: Independent Component Analysis

Topics:

ICA on mixtures of audio files

Lab 2: Inference in Graphical Models

Topics:

The sum-product algorithm
The max-sum algorithm
Medical Graph

Lab 3: Expectation Maximization and Variational Autoencoder

Topics:

Expectation Maximization
Variational Auto-Encoder

Homework 1: Exponential Families

Topics:

Probabilities,
MLE and MAP solutions
Expectation, mean and covariance
Exponential families and conjugate priors

Homework 2: Information Theory and Graphical Models

Topics:

Mutual Information, KL-divergence, entropy
Directed Graphs, Bayesian Networks, Markov Blankets

Homework 3: ICA and Markov Chains

Topics:

Conditional entropy and MI
ICA
Markov Chains and d-separation

Homework 4: Message passing

Topics:

Factor Graphs of BN
Sum-Product algorithm

Homework 5: Generative Models

Topics:

Gaussian Mixture Models
EM algorithm
Mixtures of Bernoulli

Homework 6: Sampling, Variational EM

Topics:

Rejection, Importance, Independence, Gibbs sampling
Variational EM on Mixtures of multivariate Bernoulli
Random walks

Homework 7: LDS and Causal Networks

Topics:

VEM on Linear Dynamical Systems
Causal Bayesian Networks and Simpsons's Paradox
Structural Causal Models and Truncated Factorization

Copyright

Copyright © 2019 Davide Belli.

This project is distributed under the MIT license. Please follow the UvA regulations governing Fraud and Plagiarism in case you are a student.