Matteo Tiezzi (University of Siena) Apr 15, 2020 – 11:00 AM Conference Meeting Description Many recent methods for unsupervised or self-supervised representation learning train feature extractors by maximizing an estimate of the mutual information (MI) between different views of the data. This comes with several immediate problems. For example, MI is notoriously hard to estimate, […]
Read More »Simplicity and elegance have always been incredibly useful criteria for the development of successful theories that describe natural phenomena. Variational methods frame this parsimony principles into precise mathematical statements. In this thesis we showed how we can formulate learning theories using calculus of variations. Despite the natural way in which learning problem can be formulated […]
Read More »Gabriele Ciravegna (University of Florence) Mar 4, 2020 – 11:00 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description Deep neural networks are usually considered black-boxes due to their complex internal architecture, that cannot straightforwardly provide human-understandable explanations on how they behave. Indeed, Deep Learning is still viewed with skepticism in those real-world domains […]
Read More »Giannini’s thesis A key point in the success of machine learning, and in particular deep learning, has been the availability of high-performance computing architectures allowing to process a large amount of data. However, this potentially prevents a wider application of machine learning in real world applications, where the collection of training data is often a […]
Read More »Luca Pasqualini (University of Siena) Feb 12, 2020 – 11:00 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description AlphaGo (AG) and all of its derivatives can play with superhuman strenght because they are able to predict the win-loss outcome with great accuracy. However, Go as a game is decided by a final score, […]
Read More »Alberto Rossi (University of Florence) Feb 19, 2020 – 11:00 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description This paper extends the fully recursive perceptron network (FRPN) model for vectorial inputs to include deep convolutional neural networks (CNNs) which can accept multi-dimensional inputs. A FRPN consists of a recursive layer, which, given a […]
Read More »Marco Corneli Feb 19, 2020 – 11:45 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description We consider textual interaction data involving two disjoint sets of individuals/objects. An example of such data is given by the reviews on web platforms (e.g. Amazon, TripAdvisor, etc.) where buyers comment on products/services they bought. We develop a […]
Read More »Valerio Basile (University of Turin) Feb 12, 2020 – 11:40 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description Today’s Web represents a huge repository of human knowledge, not only about facts, people, places and so on (encyclopedic knowledge), but also about everyday beliefs that average human beings are expected to hold (commonsense knowledge). […]
Read More »Lisa Graziani (University of Florence) Feb 5, 2020 – 11:00 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description Sketch generation from photos can be seen as a classic task of image to image translation with GANs. But in our work the sketch generation is addressed as an unsupervised problem based on a vectorial […]
Read More »Andrea Zugarini (University of Florence) Jan 29, 2020 – 11:00 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description We consider a scenario where an artificial agent is reading a stream of text composed of a set of narrations, and it is informed about the identity of some of the individuals that are mentioned […]
Read More »