This repository contains personal solutions to the lab assignments of the MALCOM (Machine Learning for Communication Systems) course at EURECOM (Spring 2022).
From scratch implementation of a 2-layer fully-connected neural network and test on the CIFAR-10 dataset in terms of accuracy.
Generative Adversarial Networks (GANs) for digits images generation, trained on MNIST.
Federated Learning and Clustered Federated Learning, following the approach proposed by Felix Sattler et al. in this paper.