This repository contains the solution to the projects I did for the ETH course "Probabilistic Artificial Intelligence", held by professor Andreas Krause in Autumn semester 2021. The topics are as follows:
Implementation of a Gaussian Process regression. Then applied the model to an inference problem based on space data.
Coding exercise based on the theory shown in Variational Inference for Neural Networks, implementing a simple Bayesian NN.
Implementation of a custom Bayesian optimization algorithm to an hyperparameter tuning problem.
The task was to implement an algorithm that, by practicing on a simulator, learns a control policy for a lunar lander. The method suggested is a variant of policy gradient with two additional features, namely (1) Rewards-to-go, and (2) Generalized Advantage Estimatation, both aiming at decreasing the variance of the policy gradient estimates while keeping them unbiased.