Month: March 2022

Machine Learning Models Life Cycle

In the last years, always more companies started introducing AI systems impacting their business. Machine Learning is used in e-commerce, advertisement, health, finances, driving, industry, and many other contexts. Besides the big tech players such as Google, Microsoft, Amazon and IBM, many  new startups born each year raising investments over $40.4B in 2018 (see page […]
Read More

Artificial intelligence in everyday life

Whenever Artificial Intelligence is mentioned, many imagine a dystopian future in which every single activity is carried out by machines. Furthermore, a widespread misconception is the fact that Artificial Intelligence is about the future and that it is not yet part of our lives. Artificial Intelligence, on the other hand, is applied in the most […]
Read More

Federated Learning To Predict Oxygen Needs

Cambridge’s Addenbrooke’s Hospital in collaboration with NVIDIA and 20 other hospitals around the world has used an Artificial Intelligence algorithm and Federated Learning to predict the oxygen needs of Covid-19 patients in the first days of therapy.   FEDERATED LEARNING: FEDERATED LEARNING FOR ALGORITHMIC TRAINING The AI ​​algorithm employed by Addenbrooke’s Hospital in Cambridge was […]
Read More

AI algorithm for diagnosing Covid-19 and other pathologies

The diagnosis of Covid-19 and other diseases has been accelerated thanks to an AI algorithm. Thanks to a technique called Federated Learning, the University of Cambridge International Team has developed an AI algorithm that can be trained using the datasets of other hospitals and foreign facilities. In this way, the diagnostics of Covid-19 and other […]
Read More
ICAPS2021-LOGO

Online Planning for F1 Race Strategy Identification

Authors Diego Piccinotti, Amarildo Likmeta, Nicolò Brunello, Marcello Restelli Abstract Formula 1 (F1) racing is one of the most competitive racing competitions involving high-performance single-seater racing vehicles. The result of a race is determined by vehicle and driver performance, as well as by the tire and pit-stop strategy employed in the race. In this work, […]
Read More
ICML-2021-Paper-Awards

Subgaussian Importance Sampling for Off-Policy Evaluation and Learning

Authors Alberto Maria Metelli, Alessio Russo, Marcello Restelli Abstract Importance Sampling (IS) is a widely used building block for a large variety of off-policy estimation and learning algorithms. However, empirical and theoretical studies have progressively shown that vanilla IS leads to poor estimations whenever the behavioral and target policies are too dissimilar. In this paper, […]
Read More
JMLR

Safe Policy Iteration: A Monotonically Improving Approximate Policy Iteration Approach

Authors Alberto Maria Metelli, Matteo Pirotta, Daniele Calandriello, Marcello Restelli Abstract This paper presents a study of the policy improvement step that can be usefully exploited by approximate policy–iteration algorithms. When either the policy evaluation step or the policy improvement step returns an approximated result, the sequence of policies produced by policy iteration may not […]
Read More
JMLR

MushroomRL: Simplifying Reinforcement Learning Research

Authors Carlo D’Eramo, Davide Tateo, Andrea Bonarini, Marcello Restelli, Jan Peters Abstract MushroomRL is an open-source Python library developed to simplify the process of implementing and running Reinforcement Learning (RL) experiments. Compared to other available libraries, MushroomRL has been created with the purpose of providing a comprehensive and flexible framework to minimize the effort in […]
Read More