Valeria's Projects
MPI programs exploring parallel computing techniques. It implements a message-passing ring topology, benchmarks point-to-point MPI performance, and studies various computational tasks like the Jacobi solver and 3D matrix operations. The repo evaluates performance across different topologies, networks, and nodes, using OpenMPI and IntelMPI
Parallel implementation of kd-tree construction using MPI & OpenMP. Features include efficient data point generation, tree visualization, scalability tests, and performance optimizations. Aims to address challenges like false sharing & workload imbalances.
Repository for the Advanced Topics in Machine Learning practica (@ UniTS, Spring 2023)
Automates the optimization of Multi-Layer Perceptrons using Genetic Algorithms in PyTorch for classification on MNIST dataset
Implementation of Red-Black Tree class in C++
Covid-19 hospitalization trend analysis for Regione Lombardia, covering Oct 2020-Feb 2021. Utilizes data from official sources to model and predict hospitalizations using linear, GLM, and GAM approaches. A comprehensive case study by a dedicated team.
Analyzing scores from 17 major international skating events (Oct 2016-Dec 2017). This project delves into judge biases, athlete rankings based on difficult elements, and the significance of elements versus components in final rankings. Built using Python, it offers insights derived from publicly-released International Skating Union Protocols
Repository for the DL Course AA 2021-2022 @ DSSC UniTS
A ML algorithm capable of conducting an in-depth analysis of students' responses to STACK questions
An insightful collection of finite difference methods to solve elliptic PDEs. This repository provides interactive exercises, from basic grid generation to advanced boundary value problem solutions, using forward Euler methods. Dive in to explore and understand the dynamics of PDEs through hands-on tasks.
RL framework designed to simulate and optimize government taxation policies. Focused on balancing economic equality and consumption, it employs state-of-the-art RL algorithms, PPO and SAC, to navigate wealth taxation strategies across various economic scenarios, aiming to achieve optimal wealth distribution and economic efficiency.
Predicting headache occurrences using Hidden Markov Models (HMMs). The dataset comprises 296 days of binary headache records. Analysis reveals cyclical patterns, with the DTMC Model yielding the best predictions. Wiener Process, sARIMA, Bayesian Normal Mixtures, DTMC, and Categorical Mixtures models are also reported.
Python-based Information Retrieval system leveraging the BIM probabilistic model. Features include handling free-form text queries, relevance & pseudo-relevance feedback. Performance is rigorously evaluated using metrics like precision/recall, mean average precision, and R-precision. Utilizes the standard Cranfield dataset from aerodynamics.
MIT Deep Learning Book in PDF format (complete and parts) by Ian Goodfellow, Yoshua Bengio and Aaron Courville
The simplest, fastest repository for training/finetuning medium-sized GPTs.
Advanced RL algorithms for two simplified versions of chess. Shortest Path finds the minimal moves between two cells based on piece capabilities. Capture Pieces trains against random opponents aiming for maximal captures in set moves. Features Deep Q-Learning, Policy Iteration, TD and more.
Probabilistic Machine Learning course lab @UNITS
Curated repository for all the teaching materials (courses, seminars, etc.) provided by the lab and its members.
Advanced ML analysis of Italo Svevo's epistolary corpus. Utilizing Latent Dirichlet Allocation (LDA) for topic modeling and sentiment analysis tools. Dive deep into the predominant themes and sentiments in Svevo's letters, revealing the power of ML in literary exploration.
Whatsapp chatbot API using generative pretraining transformers