1. Speaker: Sergio Orlandini (CINECA), Giuseppe Fiameni (NVIDIA), Andrea Pilzer (NVIDIA) 2-4 maggio 2023
Title: High Performance Deep Learning with GPUs
3CFU
Program: https://docs.google.com/document/d/1p0JLRtpN_p5PQtn_id8e8az5Fi17UGoIxgLqmJW2y7s/edit?usp=sharing
Abstract: Deep Learning has been the most significant breakthrough in the past ten years in the field of computer vision, natural language processing and machine learning. It has radically changed the research methodology towards a data-oriented approach, in which learning involves all steps of the prediction pipeline from feature extraction to classification. While research efforts have concentrated on the design of effective feature extraction and prediction architectures, computation has moved from CPU-only approaches to the dominant use of GPUs and massively parallel devices, empowered by large-scale and highly dimensional datasets. Thus the goal of this course is to introduce parallel architectures, accelerated systems, code parallelism, basic concepts and model training on large scale systems.
2. Speaker: Luca Becchetti e Francesco Pasquale, 12-16 Giugno 2023
Title: Spectral graph theory and random walks: connections and applications
3CFU
Abstract. The spectral properties of graphs are of paramount importance. While important properties of graphs are reflected in the properties of their spectra, fundamental connections between spectral and topological properties of graphs are at the core of some successful approaches in a number of problems, e.g., graph partitioning or community detection. At the same time, graph spectra are related to key properties of random walks in graphs. These, in turn, describe key aspects of some fundamental dynamics on graphs. The goal of this course is to introduce students to these topics, providing an overview of key notions, some reference applications, and how spectral properties of graphs can be investigated from different perspectives, leading to insights into applications of interest both in computer science and other areas of research.
3. Speaker: Amin Mantrach (Amazon) and James Thorne (KAIST), 3-13 Luglio 2023
Title: AI in the Wild & Knowledge Intensive NLP
3CFU
AI-based systems are more and more present in industrial setting powering production systems as recommender systems, search engines, autonomous driving vehicles, and control systems. In this course, we will dive deep on what are the essential steps to build and maintain such as system. We will cover model training, model evaluation, model deployment, automatic model refresh, monitoring systems, etc.
4. Speaker: Vangelis Markakis (University of Athens) and Ivan Visconti (University of Salerno), 25-29 September 2023
Title: Crypto and incentive-based Mechanisms for Blockchain Technology
3CFU
Abstract: The Blockchain Era started with the cryptocurrency Bitcoin that Nakamoto designed by crucially and cleverly using digital signatures and cryptographic hash functions. Nowadays mainstream blockchains and their applications strongly rely on more advanced cryptographic tools that in the past were considered of theoretical interest only. The goal of this course is to overview the various cryptographic notions that have played a key-role for the success of blockchains and their applications.
5. Speaker: Daniele Oxoli, Maria Antonia Brovelli, Marco Gianinetto, Andrea Monti Guarnieri, Michelle Lavagna
Title: The Copernicus Green Revolution for Sustainable Development.
5CFU
Non mandatory