PAI_Projects

Probabilistic Artificial Intelligence

Gaussian Process Regression

We implemented Gaussian Process regression from scratch and used it for the computation and estimation of the model posterior. Finally, we applied our method to a dataset based on space data.

Bayesian Neural Network

We implemented a Bayesian NN and applyied it on the Rotated MNIST and Fashion MNIST datasets.

Bayesian Optimization

We implemented Bayesian Optimisation to find the best hyperparameters of a model. In particular, the goal was to determine the value of a model hyperparameter that maximizes the validation accuracy subject to a constraint on the average prediction speed.

Actor Critic Reinforcement Learning

We implemented a Reinforcement Learning Algorithm based on Actor-Critic that solves the lunar lander environment.