Program‎ > ‎

Plenary sessions


Convexity, Sparsity, Nullity and all that... in Machine Learning

by Dr. Hamid Krim (North Carolina State University). Monday, March 9th, h.14:00. Room Giolitti.
High dimensional data exhibit distinct properties compared to its low dimensional counterpart; this causes a common performance decrease and a formidable computational cost increase of traditional approaches. Novel methodologies are therefore needed to characterize data in high dimensional spaces.
Considering the parsimonious degrees of freedom of high dimensional data compared to its dimensionality, we study the union-of-subspaces (UoS) model, as a generalization of the linear subspace model. The UoS model preserves the simplicity of the linear subspace model, and enjoys the additional ability to address nonlinear data. We show a sufficient condition to use minimization to reveal the underlying UoS structure, and further propose a bi-sparsity model (RoSure) as an effective algorithm, to recover the given data characterized by the UoS model from errors/corruptions.
As an interesting twist on the related problem of Dictionary Learning Problem, we discuss the sparse null space problem (SNS). Based on linear equality constraint, it first appeared in 1986 and has since inspired results, such as sparse basis pursuit, we investigate its relation to the analysis dictionary learning problem, and show that the SNS problem plays a central role, and may naturally be exploited to solve dictionary learning problems.
Substantiating examples are provided, and the application and performance of these approaches are demonstrated on a wide range of problems, such as face clustering and video segmentation.

Biography


Dr. Hamid Krim (ahk@ncsu.edu) received his BSc., MSc. and Ph.D. in Electrical Engineering. He was a Member of Technical Staff at AT&T Bell Labs, where he has conducted R&D in the areas of telephony and digital communication systems/subsystems. Following an NSF postdoctoral fellowship at Foreign Centers of Excellence, LSS/University of Orsay, Paris, France, he joined the Laboratory for Information and Decision Systems, MIT, Cambridge, MA as a Research Scientist and where he was performing and supervising research.  
He is presently Professor of Electrical Engineering in the ECE Department, North Carolina State University, Raleigh, leading the Vision, Information and Statistical Signal Theories and Applications group. His research interests are in statistical signal and image analysis and mathematical modeling with a keen emphasis on applied problems in classification and recognition using geometric and topological tools. He has served and is currently serving on the IEEE editorial board of SP, and the TCs of SPTM and Big Data Initiative, as well as an AE of the new IEEE Transactions on SP on Information Processing on Networks, and of the IEEE SP Magazine. He is also one of the 2015-2016 Distinguished Lecturers of the IEEE SP Society.


Can Signal Processing Help You Find Your Way?

by Dr. Fabio Dovis (Politecnico di Torino - DET). Tuesday, March 10th, h.9:30. Room Giolitti.
In the past few years, the interest towards satellite navigation systems significantly increased. New Global Navigation Satellite Systems (GNSS) in addition to the current American Global Positioning System(GPS) and the Russian GLONASS are being developed leading the GNSS world towards a new era where a wider plethora of services and signals will be available for the end user. In parallel to the deployment of new systems providing new signals for positioning, the receiver technology is taking advantage of the software-defined-radio concepts to design software receivers that are in perspective replacing GNSS equipment making use of dedicated chipsets and platforms.
This new flexibility in the receiver architecture design, among the several advantages, has opened opportunities for the implementation of advanced signal processing algorithms able to improve the quality of the estimation of the position.
The majority of the positioning services in fact, needs high accuracy and reliability, and due to the very low received signal strength, this can only be granted by advanced receivers able to deal with propagation effects in the dispersive layers of the atmosphere, as well as to deal with intentional and unintentional interference that might be present in the operational environment.
On the other hand, the range of applications of satellite-based positioning is widening encompassing nowadays, fields that are different from the classical transportation and navigation sectors: remote sensing, smart grids, space weather are just few examples of fields in which the use of GNSS-based techniques is becoming popular and creating innovation.
This presentation will show how signal processing techniques developed in different frameworks are cross-fertilizing the satellite navigation field both for improving the positioning performance, as well as a mean to exploit the GNSS signals for new applications and services.

Biography

Fabio Dovis
Dr. Fabio Dovis
 (fabio.dobis@polito.it) is associate professor at the Department of Electronics and Telecommunications of Politecnico di Torino as a member of the Navigation Signal Analysis and Simulation (NavSAS) group. He obtained in 2000 his PhD in Electronics and Communications eng. from Politecnico di Torino. His research interests cover the design of GPS and Galileo receivers and advanced signal processing for interference and multipath detection and mitigation. He has a relevant experience in European projects in satellite navigation as well as cooperation with industries and research centers. Currently he supports the European GNSS Agency and the Galileo Unit of the European Commission, as external technical advisor for monitoring on-going research projects. In June 2011 he has been appointed as member of the Mission Evolution Advisory Group of the European Commission, the group of 23 experts having the task to propose and evaluate evolutions of the mission objectives for the European satellite navigation programmes Galileo and EGNOS. In 2007 he has been included in the list of the “50+ people to watch” by the GPS World magazine, a list of the people who provided significant contributions to the development of the navigation field.
Plenary presentations download