My personal solution to the laboratories for the Advanced Statistical Inference course at EURECOM, Sophia Antipolis.
In this notebooks I explored the major concepts of Bayesian Machine Learning, applying Bayesian statistical inference both to regression and classification problems. Moreover I also had the chance to explore the power of Gaussian Processes, having to implement a simple version of this extremely powerful technique.
Alongside with those, I also explored some other common techniques in Machine Learning such as clustering and dimensionality reduction
All those work were largely inspired by the two following books:
In the assignment folder I developed a classification algorithm based on statistical inference and tried to combine it with deep learning Convolutional Neural Network techniques. My classifier was then applied on the MNIST and CIFAR10 dataset, reaching respectively 57.53% and 29.76% accuracy.
More details about the requirements of the assignment can be found here.