Welcome!

This is the official web site of Siena Artificial Intelligence Laboratory. The focus of our research is on machine learning. In the last few years, we’ve been mainly involved in the conception of new theories of learning in structured domains and in their applications to pattern recognition and mining the web. We are also interested […]

Read More »

DeepHarvest

Do you want to challenge a Deep Reinforcement Learning based algorithm? If yes, then DeepHarvest is a very good opportunity! You can find all the information and play an ONLINE DEMO here: https://sailab.diism.unisi.it/deepharvest

Read More »

[ Sept 13th 2018] LabMeeting: Emotion recognition from audio

Lisa Graziani (DIISM, Universities of Florence and Siena) Sept 13, 2018 – 9:30 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description Emotion recognition from audio is a widely studied topic, but is still very challenging because is not entirely clear which features are effective for the recognition task. Moreover voice features keep on […]

Read More »

Visual Attention Modeling

ONLINE DEMO >> Computational models of visual attention are at the crossroad of disciplines like cognitive science, computational neuroscience, and computer vision. When eye-tracking devices are not a viable option, models of human attention can be used to predict fixations.  Not only humans are correlated in terms of the locations they fixate, but they also agree […]

Read More »

[ Sept 6th 2018] LabMeeting: Adversarial Reprogramming of Neural Networks

Matteo Tiezzi (DIISM, University of Siena) Sept 6, 2018 – 9:30 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description Adversarial examples are defined as “inputs to machine learning models that an attacker has intentionally designed to cause the model to make a mistake”. Indeed, in the computer vision scenario it has been shown […]

Read More »

[Jul 23th 2018] On stochastic gradient descent, flatness and generalization – Seminar by Prof. Yoshua Bengio

Prof. Yoshua Bengio (University of Montreal, Department of Computer Science and Operations Research (DIRO) ) Jul 23, 2018 – 11:00 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description The traditional Machine Learning picture is that optimization and generalization are neatly separated aspects. That makes theory easier to handle, separately, but unfortunately this is not the case. Stochastic […]

Read More »

First Place at Hackathon Infinity

Three of our PhD students (Giuseppe Marra, Dario Zanca, and Andrea Zugarini, together with Giovanni Ciccone) won the Hackathon Mediaset Infinity, held in Rimini at Web Marketing Festival 2018. The goal of the challenge was to predict churn rate of users registered in infinitytv platform. Ten teams, composed of both students and professionals from Statistics […]

Read More »

[ Jun 21st 2018] Lab Meeting: Loss functions generation by means of fuzzy aggregators

Francesco Giannini (DIISM, University of Siena) Jun 21, 2018 – 9:30 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description In a learning from constraints problem, the prior knowledge can be expressed by logical formulas and then converted into real valued functions. Hence the satisfaction of a certain formula F can be enforced by […]

Read More »

[Jun 7th 2018] LabMeeting: Probabilistic Soft Logic

Giuseppe Marra (DIISM, University of Siena) Jun 7, 2018 – 9:30 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description A fundamental challenge in developing high-impact machine learning technologies is balancing the need to model rich, structured domains with the ability to scale to big data. Many important problem areas are both richly structured […]

Read More »

[May 31st 2018] LabMeeting: Depth growing neural networks

Vincenzo Laveglia (DIISM, University of Siena) May 31, 2018 – 9:30 AM DIISM, Artificial Intelligence laboratory (room 201), Siena SI Description Being able to train neural networks (shallow or deep) requires the ability to identify the right hyper-parameters for the model. Some of these hyper-parameters are more related to the training procedure (epochs, learning rate, […]

Read More »