4th Italian Workshop on Machine Learning and Data Mining (MLDM.it 2015)

Following the success of the first three editions of the Italian Workshop on Machine Learning and Data Mining at the AI*IA Symposiums and AI*IA Conferences on Artificial Intelligence, this workshop aims at bringing together researchers actively involved in the fields of machine learning, data mining, pattern recognition, and knowledge discovery.

During the workshop, researchers will have the opportunity to present their recent results and discuss problems and challenges relevant to the community. Following the tradition of MLDM.it, presentations are upon invitation. Attendance to MLDM is open to all the AI*IA Conference participants.

The meeting is organized by the AI*IA Working Group on Machine Learning and Data Mining whose general goal is to promote Italian scientific activities in the field of machine learning and data mining, and foster collaborations between research groups.

Workshop Day: September 22nd, 2015

Program

MLDM.it will be held in Main Room - First Floor, Palazzo Trotti Mosti (Trotti Mosti Palace): Corso Ercole I d'Este, 37 - Building Plan

The program is downloadable here.

  22 September 2015  
8.50-9.00 Opening
9.00-9.10 Welcome
Section I Complex Data Analysis
9.10-9.30 Mining Time-Adaptive Interpolative Clusters: Linking Interpolation to Summarization in Geophysical Data Streams Annalisa Appice, Donato Malerba
9.30-9.50 Visual Detection of small groups and leaders in crowd through structured learning Simone Calderara
9.50-10.10 Real-Time Topic Detection in Twitter Salvatore Gaglio, Giuseppe Lo Re, Marco Morana
10.10-10.30 Choice of Training-Validation partitions impacts predictive performance Alessandro Zandonà, Marco Chierici, Giuseppe Jurman, Cesare Furlanello
Coffee break
Section II Deep Learning
11.00-11.20 Two approaches for Recurrent Neural Networks Pre-training Luca Pasa, Alessandro Sperduti, Alberto Testolin
11.20-11.40 On the Complexity of Neural Network Classifiers: A Comparison between Shallow and Deep Architectures Monica Bianchini, Franco Scarselli
11.40-12.00 Deep Reservoir Computing Claudio Gallicchio, Alessio Micheli
12.00-12.20 Tweaking Sum-Product Network Structure Learning Antonio Vergari, Nicola Di Mauro, Floriana Esposito
Lunch break
Section III Learning Models and Analysis (I)
14.30-14.50 Distributed Probabilistic Logic Learning Elena Bellodi, Giuseppe Cota, Evelina Lamma, Fabrizio Riguzzi, Riccardo Zese
14.50-15.10 Truly scalable latent locally linear SVMs Ilja Kuzborskij, Barbara Caputo
15.10-15.30 A model of online learning as a Linear Quadratic Gaussian (LQG) optimal control problem with random matrices Giorgio Gnecco, Alberto Bemporad, Marco Gori, Rita Morisi, and Marcello Sanguineti
15.30-15.50 Grammatical Inference for Structural Knowledge Extraction Pietro Cottone, Salvatore Gaglio, Giuseppe Lo Re, and Marco Ortolani
Coffee break
Section IV Learning Models and Analysis (II)
16.30-16.50 RNAsynth: a graph kernel approach to learn constraints for RNA inverse folding Fabrizio Costa
16.50-17.10 Transducing Sentences to Syntactic Feature Vectors: an Alternative Way to “Parse”? Fabio Massimo Zanzotto, Lorenzo Dell’Arciprete, Lorenzo Ferrone
17.10-17.30 Conceptual change as a cognitive phase transition Lorenza Saitta
17.30-17.45 Model complexities of shallow neural networks for the approximation of input-output mappings with large variations Marcello Sanguineti, Vera Kurkova
17.45-18.00 Discussion and Conclusions: The future of MLDM

Organising Committee

Alessio Micheli
University of Pisa

Giorgio Valentini
University of Milano

Advisory board

Nicolò Cesa-Bianchi
Paolo Frasconi
Alessio Micheli
Marcello Pelillo
Giorgio Valentini
Alessandro Verri