CBCL Home

CBCL People
home people publications courses news video research events code/data jobs

The IIT@MIT lab was founded from an agreement between the Massachusetts Institute of Technology (MIT) and the Istituto Italiano di Tecnologia (IIT). The scientific objective is to develop novel learning and perception technologies – algorithms for learning, especially in the visual perception domain, that are inspired by the neuroscience of sensory systems and are developed within the rapidly growing theory of computational learning. The ultimate goal of this research is to design artificial systems that mimic the remarkable ability of the primate brain to learn from experience and to interpret visual scenes.

Publications

Regularization Predicts While Discovering Taxonomy. Y. Mroueh, T. Poggio and L. Rosasco (2011); CVPR Workshop on Fine-Grained Visual Categorization (FGVC) (also available as MIT-CSAIL-TR-2011-029/CBCL-299, Massachusetts Institute of Technology, Cambridge, MA).

Multi-class Learning: Simplex Coding and Relaxation Error. Y. Mroueh, T. Poggio, L. Rosasco and J.J. Slotine (2011); Oberwolfach Mini-Workshop on Mathematics of Machine Learning.

Multiscale Geometric Methods for Estimating Intrinsic Dimension. A.V. Little, M. Maggioni, L. Rosasco (2011); Proceedings of SampTA 2011.

Some recent advances in multiscale geometric analysis of point clouds. G. Chen, A.V. Little, M. Maggioni and L. Rosasco (2011); Wavelets and Multiscale Analysis: Theory and Applications.

Multi-Output Learning via Spectral Filtering. L. Baldassarre, L., A. Barla, L. Rosasco and A. Verri (2011); accepted for publication in Machine Learning. (aslo available as MIT-CSAIL-TR-2011-004/CBCL-296).

Nonparametric Sparsity and Regularization. S. Mosci, L. Rosasco, M. Santoro, A. Verri and S. Villa (2011); submitted for publication to Journal of Machine Learning Research (available online as MIT-CSAIL-TR-2011-041/CBCL-303, Massachusetts Institute of Technology, Cambridge, MA).

Kernels for Vector-Valued Functions: a Review. M. Alvarez, N. Lawrence and L. Rosasco (2011); MIT-CSAIL-TR-2011-033/CBCL-301 Massachusetts Institute of Technology, Cambridge, MA.

Online Learning, Stability, and Stochastic Gradient Descent. T. Poggio, S. Voinea and L. Rosasco (2011); Cornell University Library, arXiv:1105.4701v2 [cs.LG].

Consistency of learning algorithms using Attouch-Wets convergence. S. Villa, L. Rosasco, S. Mosci and A. Verri (2010); to be published in Optimization (currently available with DOI: 10.1080/02331934.2010.511671).

Learning Generic Invariances in Object Recognition: Translation and Scale. J.Z. Leibo, J. Mutch, L. Rosasco, S. Ullman, and T. Poggio (2010); MIT-CSAIL-TR-2010-061/CBCL-294, Massachusetts Institute of Technology, Cambridge, MA, December.

Neurons That Confuse Mirror-Symmetric Object Views. J. Mutch, J.Z. Leibo, S. Smale, L. Rosasco, and T. Poggio (2010); MIT-CSAIL-TR-2010-062/CBCL-295, Massachusetts Institute of Technology, Cambridge, MA, December.

Spectral Regularization for Support Estimation. E. De Vito, L. Rosasco and A. Toigo (2010); Advances in Neural Information Processing Systems (NIPS).

A primal-dual algorithm for group sparse regularization with overlapping groups. S. Mosci, S. Villa, and L. Rosasco (2010); Advances in Neural Information Processing Systems (NIPS).

Solving structured sparsity regularization with proximal methods. S. Mosci, L. Rosasco, M. Santoro, A. Verri, and S. Villa (2010); in Machine Learning and Knowledge Discovery in Databases, volume 6322 of Lecture Notes in Computer Science (Springer).

A regularization approach to nonlinear variable selection. L. Rosasco, M. Santoro, S. Mosci, A. Verri, and Silvia Villa (2010); in Proceedings of The Thirteenth International Conference on Artificial Intelligence and Statistics (AISTATS), volume 9 of JMLR W&CP.