Fabio Aiolli

Assistant Professor

Research and Publications

Main Research Interests:

SVMs and Kernel-based Algorithms
Preference Learning
Neural Networks and Kernels for Structures
Transfer Learning
On-line Algorithms
Tangent Distance
ML applications in Biology and Paleography
Recommender Systems

News:

2012 MILLION SONGS DATASET Challenge CODE is here!
2012 WINNER of the MILLION SONGS DATASET challenge at Kaggle (http://www.kaggle.com/c/msdchallenge)

2011 UTL CHALLENGE (phase I) BEST PAPER AWARD
2011 WINNER of the first phase of the UNSUPERVISED AND TRANSFER LEARNING challenge (http://www.causality.inf.ethz.ch/unsupervised-learning.php)

Publications:

International Journals and Conference Proceedings
M. Donini, F. Aiolli	Learning Deep Kernels in the space of dot-product polynomials Machine Learning (2016)
M. Ciman, M. Donini, O. Gaggi, F. Aiolli	Stairstep recognition and counting in a serious game for increasing users' phisical activity Personal and Ubiquitous Computing (2016)
L. Oneto, N. Navarin, M. Donini, F. Aiolli, D. Anguita	Advanced in learning with kernels: theory and practice in a world of growing constraints ESANN 2016, Bruges (Belgium)
L. Oneto, N. Navarin, M. Donini, A. Sperduti, F. Aiolli, D. Anguita	Measuring the expressiveness of graph kernels through the Rademacher complexity ESANN 2016, Bruges (Belgium)
M. Polato, F.Aiolli	Kernel based Collaborative Filtering for very large scale top-N item recommendation ESANN 2016, Bruges (Belgium)
F. Aiolli, M. Donini, N. Navarin, A. Sperduti	Multiple Graph Kernel Learning SSCI 2015, Cape Town (South Africa)
F. Aiolli, M. Donini	EasyMKL: a scalable multiple kernel learning algorithm Neurocomputing (2015)
V. Bolon Canedo, M. Donini, F. Aiolli	Feature and kernel learning Proceedings of ESANN 2015 (Bruges, BE)
F. Aiolli, M. Ciman, M. Donini, O. Gaggi	ClimbTheWorld: Real-time stairstep counting to increase physical activity Proceedings of the 11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, 2014
F. Aiolli	Convex AUC Optimization for Top-N Recommendation with Implicit Feedback Proceedings of the ACM RecSys 2014 (Sylicon Valley, USA)
F. Aiolli, G. Da San Martino, A. Sperduti	An Efficient Topological Distance-based Tree Kernel IEEE Transaction on Neural Networks and Learning Systems (2014)
F. Aiolli, M. Donini	Learning Anisotropic RBF Kernels Proceedings of ICANN 2014 (Hamburg, GE)
F. Aiolli, M. Donini	Easy Multiple Kernel Learning Proceedings of ESANN 2014 (Bruges, BE)
F. Aiolli	Efficient Top-N R ecommendation for Very Large Scale Binary Rated Datasets Proceedings of the ACM RecSys 2013 (Hong Kong, CH)
F. Aiolli, M. Donini, E. Poletti, E. Grisan	Stacking Models for Efficient Annotation of MR Volumes Proceedings of MEDICON 2013 (Siviglia, SP)
F. Aiolli	A Preliminary Study on a Recommender System for the Million Song Dataset Challenge Proceedings of IIR 2013, 73-83 (Pisa, IT)
F. Aiolli	Transfer Learning by Kernel Meta-Learning Journal of Machine Learning Research W&CP, 27:81-95, 2012 (Pascal2 Best Challenge Paper Award)
T. Sanavia, F. Aiolli, G. Da San Martino, A. Bisognin, B. Di Camillo	Improving biomarker list stability by integration of biological knowledge in the learning process BMC Bioinformatics, vol. 13(4), 2012.
F. Aiolli, G. Da San Martino, A. Sperduti	Extending Tree Kernels with Topological Information Proceedings of the ICANN Conference, 2011: 142-149
F. Aiolli, A. Burattin, A. Sperduti	A Business Process Metric Based on the Alpha Algorithm Relations Business Process Management Workshops (1) 2011, 141-146
F. Aiolli, M. Giollo	A Study on the Writer Identification Task for Paleographic Document Analysis Proceedings of the 11th IASTED International Conference on Artificial Intelligence and Applications (AIA), 2011
F. Aiolli, G. Da San Martino, A. Sperduti	A New Tree Kernel Based on SOM-SD Proceedings of the International Conference on Artificial Neural Networks (ICANN), vol. 2, p. 48-58, 2010
F. Aiolli, C.E. Palazzi	Enhancing Artificial Intelligence on a Real Mobile Game International Journal of Computer Games Technology (special issue on Artificial Intelligence for Computer Games), 2009.
F. Aiolli, G. Da San Martino, M. Hagenbuchner, A. Sperduti	Learning Nonsparse Kernels by Self-Organizing Maps for Structured Data IEEE Transactions on Neural Networks, vol. 20, p. 1938-1949, 2009.
F. Aiolli, M. De Filippo, A. Sperduti	Application of the Preference Learning Model to a Human Resource Selection Task Proceedings of the Symposium on Computational Intelligence and Data Mining (CIDM), 2009
F. Aiolli, G. da San Martino, A. Sperduti	Route Kernels for Trees Proceedings of the International Conference on Machine Learning (ICML), 2009
F. Aiolli, A. Sperduti	Supervised Learning as Preference Optimization Proceedings of the European Symposium on Artificial Neural Networks (ESANN), 2009
.F. Aiolli, R. Cardin, F. Sebastiani, A. Sperduti	Preferential Text Classification: Learning Algorithms and Evaluation Measures Information Retrieval, 2008.
F. Aiolli, G. Da San Martino, A. Sperduti	A Kernel Method for the Optimization of the Margin Distribution Proceedings of the ICANN Conference, Praga 2008.
F. Aiolli, C. E. Palazzi	Enhancing Artificial Intelligence in Games by Learning the Opponent's Playing Style Proceedings of the IFIP-ECS Conference, Milano 2008
F. Aiolli, F. Sebastiani, A. Sperduti	Preference Learning for Category-Ranking based Interactive Text Categorization Proceedings of the IJCNN Conference, Orlando (USA) 2007
F. Aiolli, G. Da San Martino, M. Hagenbuchner, A. Sperduti	"Kernelized" Self Orginizing Maps for Structured Data Proceedings of the ESANN Conference, Bruges 2007
F. Aiolli, G. Da San Martino, A. Moschitti, A. Sperduti	Efficient Kernel-based Learning for Trees Proceedings of the CIDM Conference, Honolulu Hawaii 2007
F. Aiolli, G. Da San Martino, A. Moschitti, A. Sperduti	Fast On-line Kernel Learning for Trees Proceedings of the ICDM Conference, Hong Kong 2006
F. Aiolli, A. Sperduti	Multiclass Classification with Multi-Prototype Support Vector Machines Journal of Machine Learning Research, 6(May) 2005 - pages 817-850.
F. Aiolli	A Preference Model for Structured Supervised Learning Tasks Proceedings of the ICDM Conference, New Orleans (Louisiana, USA) 2005 (Long Version)
F. Aiolli, A. Sperduti	Learning Preferences for Multiclass Problems Proceedings of the NIPS Conference, Vancouver Canada 2004.
F. Aiolli, A. Sperduti	Multi-prototype Support Vector Machines Proceedings of the IJCAI Conference, Acapulco Mexico 2003.
F. Aiolli, F. Portera, A. Sperduti	Speeding up the solution of multi-label problems with Support Vector Machines Supplementary Proceedings of the ICANN/ICONIP Turkey, 2003.
F. Aiolli, A. Sperduti	A re-weighting strategy for improving margins AI Journal, 2002 (137) - pages 197-216.
F. Aiolli, A. Sperduti	An Efficient SMO-like algorithm for Multiclass SVM Proceedings of the NNSP, 2002.
F. Aiolli, A. Sperduti	A simple additive re-weighting scheme for improving margins Proceedings of the IJCAI Conference, Seattle USA 2001.

Contributions in Books
F.Aiolli	Transfer Learning by Kernel Meta-Learning In Guyon I., Dror G., Lemaire V., Taylor G., Sylver D., Unsupervised and Transfer Learning. Challenges in Machine Learning series (vol.7) of Microtome. In Press
F. Aiolli, A. Sperduti	A Preference Optimization Based Framework for Supervised Learning Problems In Furnkranz J., Hullermeier E., Preference Learning, pag 19-42, Springer, 2010
F. Aiolli, A. Ciula	A Case Study on the System for Paleographic Inspections (SPI): Challenges and New Develepments In Masulli F., Micheli A., Sperduti A. Computational Intelligence and Bioengineering, p. 53-66, 2009
F. Aiolli, G. Da San Martino, M. Hagenbuchner, A. Sperduti	Self Organizing Maps for Structured Domains: Theory, Models, and Learning of Kernels In Innovation in Neural Information Paradigms and Applications. vol. 247, p. 9-42, 2009

Workshops
F.Aiolli	A Preliminary Study of a Recommender System for the Million Songs Dataset Challenge Proceedings of the ECAI Workshop on Preference Learning: Problems and Application in AI, Montpellier (FR), 2012
F. Aiolli, A. Sperduti	Supervised Learning as Preference Optimization: Recent Applications Proceedings of the ECML/PKDD Workshop on Preference Learning, Antwerp (BE), 2008

(Past) Events:

SPECIAL SESSION on "Learning (with) Preferences" at ESANN (European Symposium on Artificial Neural Networks) 2009
Organizers: Fabio Aiolli, Alessandro Sperduti (Univ. degli Studi di Padova, Italy)

Other Publications:

BIZON-IST Consortium, BIZON business requirements
E-COMM-LINE 2002 Conference in Bucharest, 2002.
BIZON-IST Consortium, Semi-Automatic Annotation in E-business Applications
E-COMM-LINE 2002 Conference in Bucharest, 2002.
F. Aiolli, M. Simi, D. Sona, A. Sperduti, A. Starita, G. Zaccagnini, SPI: A System for Paleographic Inspections, Notiziario AI*IA, N. 4, December 1999.
F. Aiolli, Sviluppo di un sistema basato sulla Distanza Tangente per l'Analisi Morfologica di Caratteri Paleografici, Master Thesis, February 1999.

PhD Thesis

Large Margin Multiclass Learning: Models and Algorithms. Dept. of. Computer Science, University of Pisa

Master Thesis

Sviluppo di un sistema basato sulla distanza tangente per l'analisi morfologica di caratteri paleografici (A Tangent Distance based system for morphological analysis of paleographic characters), Dept. of. Computer Science, University of Pisa

In my master thesis I applied Tangent Distance based algorithms to the Paleographic field with very promising results.
Personally, I implemented SPI (A System for Palegraphic Inspections), an integrated system that applies innovative segmentation and supervised learning techniques to the problem of paleographic document retrieval.

Last modified:
Nov 25, 2016.