Riccardo Cadei's Projects
Discussion of advantages and disadvantages of AdaHessian, a state-of-the-art Second Order Methods over First Order Methods on a Non-Convex Optimization Problem (digits classification on MNIST database using ResNet18). - @ EPFL
[CVPR22] Towards Robust and Adaptive Motion Forecasting: A Causal Representation Perspective
A statistical framework for estimating the climate penalty on soot pollution.
The Causal Rule Ensemble Method
Collection of simulations and real world experiments on Interpretable Inference of Heterogeneous Treatment Effects by Causal Rule Ensemble
Mini Deep Learning framework to create and train simple Neural Network (i.e. Multi Layer Perceptron) through Back-Propagation Algorithm using only PyTorch's tensor definition and elementary mathematical operations.
Comparing couple of images from MNIST dataset using DCNN.
Classification of satellite images among Residential, Industrial and No Available Rooftop Area class using Deep Learning - LESO-PB Lab @ EPFL
Multiโlabel vertex classification on a proteinโprotein interaction network generating a powerful embedding of its graph - @ Oracle Labs, Politecnico di Milano
A classification problem on a big physical dataset simulated by the ATLAS experiment from CERN - @ EPFL
Teaching to an agent to play the Lunar Lander game from OpenAI Gym using REINFORCE.
Deploying a 5G Network in a country
Customer churn classification - @ Politecnico di Milano
Detecting available rooftop area from satellite images to install photovoltaic panels - LESO-PB Lab @ EPFL
A self-supervised learning at test time approach for robust and adaptable diagnosis of pneumonia on chest X-rays data.
Python implementation of Causal Rule Ensemble algorithm.
Introduction
Personal Portfolio
Robust and Efficient Journey Planning for Switzerland - @ EPFL