Unsupervised learning from continuous visual streams is a challenging problem that cannot be naturally and efficiently managed in the classic batch-mode setting of computation. The information stream must be carefully processed accordingly to an appropriate spatio-temporal distribution of the visual data, while most approaches of learning commonly assume uniform probability density. In this paper we focus on unsupervised learning for transferring visual information in a truly online setting by using a computational model that is inspired to the principle of least action in physics. The maximization of the mutual information is carried out by a temporal process which yields online estimation of the entropy terms. The model, which is based on second-order differential equations, maximizes the information transfer from the input to a discrete space of symbols related to the visual features of the input, whose computation is supported by hidden neurons. In order to better structure the input probability distribution, we use a human-like focus of attention model that, coherently with the information maximization model, is also based on second-order differential equations. We provide experimental results to support the theory by showing that the spatio-temporal filtering induced by the focus of attention allows the system to globally transfer more information from the input stream over the focused areas and, in some contexts, over the whole frames with respect to the unfiltered case that yields uniform probability distributions.
Matteo Tiezzi (University of Siena)
Matteo Tiezzi - PhD Student @ SAILab I am a PhD candidate in Information Engineering at University of Siena. I am generally interested in Artificial Intelligence and other connected areas of computer science. My recent contributes are related to both the application of Deep Learning models to computer vision and video analysis, and fundational studies regarding novel learning algorithms and neural models. I am particularly interested in the emerging field of non-euclidean Deep Learning, in particular in Graph Neural Networks.
Stefano Melacci (University of Siena)
Alessandro Betti (University of Siena)
Marco Maggini (University of Siena)
Marco Gori (University of Siena)
More from the Same Authors
2020 : Learning to Identify Drilling Defects in TurbineBlades with Single Stage Detectors »
Andrea Panizza · Szymon Tomasz Stefanek · Stefano Melacci · Giacomo Veneri · Marco Gori
2022 Poster: Concept Embedding Models »
Mateo Espinosa Zarlenga · Pietro Barbiero · Gabriele Ciravegna · Giuseppe Marra · Francesco Giannini · Michelangelo Diligenti · Zohreh Shams · Frederic Precioso · Stefano Melacci · Adrian Weller · Pietro Lió · Mateja Jamnik