pascalesser/machine-learning

Machine Learning Courses 2017/2018 - MSc Artificial Intelligence @ UvA

Machine Learning

Description

Homeworks and labs of the Machine Learning 1 and Machine Learning 2 courses of the MSc in Artificial Intelligence at the University of Amsterdam. Homeworks are individual, while labs are joint work with Gabriele Bani.

Machine Learning 1

Homeworks

• Basic Linear Algebra, Matrix Calculus and Probability Theory: Problem statement and Solution
• Maximum a posteriori Solution for Linear Regression, Probability Distributions, Likelihoods, and Estimators: Problem statement - Solution
• Naive Bayes Classification and Multi-class Logistic Regression: Problem statement - Solution
• Constrained Optimization, Kernel Methods and Neural Networks: Problem statement - Solution
• Mixture Models and PCA: Problem statement - Solution

Labs

Polynomial Regression and Bayesian Regression

Problem statement and Solution

Polynomial Regression	Bayesian Regression

Multi-Class Logistic Regression and Multilayer Perceptrons

Problem statement and Solution

Learned Weights for single-layer MLP	Learned Weights for Logistic Regression

Gaussian Processes and Support Vector Machines

Problem statement and Solution

Gaussian Process Regression

SVM Classification

Machine Learning 2

Homeworks

• Probability Distributions and Exponential Family: Problem statement - Solution
• Information Theory and Bayesian Networks: Problem statement - Solution
• Exponential Families, ICA, Graphical Models and D-Separation: Problem statement - Solution
• Bayesian Networks, Factor Graphs and Message Passing Algorithm: Problem statement - Solution
• Expectation Maximization: Problem statement - Solution
• Sampling Methods and Variational Expectation Maximization: Problem statement - Solution
• Linear Dynamical Systems and Causality: Problem statement - Solution

Labs

Independent Component Analysis

Problem statement and Solution

Original Signals	Reconstructed Signals

Inference in Graphical Models

Problem statement and Solution

Factor Graph of Interest

Expectation Maximization and Variational Autoencoder

Problem statement and Solution

Manifold Learned by VAE on MNIST

Testing

Refer to each notebook name and run Jupyter notebook with a following command:

jupyter notebook #notebook#.ipynb

Dependencies

• Numpy
• Matplotlib
• Scipy
• IPython Notebook

Copyright

Copyright © 2017-2018 Andrii Skliar.

This project is distributed under the MIT license. This was developed as part of the courses Machine Learning 1 and Machine Learning 2 taught by Rianne van den Berg, Max Welling, Patrick Forré and Joris Mooij at the University of Amsterdam. Please follow the UvA regulations governing Fraud and Plagiarism in case you are a student.