BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : Sparsity and Low Rank for Robust Social Data Analy
 tics and Networking
DTSTART:20121121T161500
DTEND:20121121T173000
DTSTAMP:20260510T202345Z
UID:f15ff65f4e92116b8f176b2779fea98eefd60a122b98d1375720a064
CATEGORIES:Conferences - Seminars
DESCRIPTION:Georgios B. Giannakis\nEndowed Chair Prof. Dept. of ECE\nDirec
 tor of Digital Technology Center University of Minnesota\, USA\nAbstract\
 nThe information explosion propelled by the advent of personal computers\,
  the Internet\, and the global-scale communications has rendered statistic
 al learning from `Big Data' increasingly important for analysis and proces
 sing. Along with data adhering to postulated models\, present in large vol
 umes of data are also those that do not – what are referred to as outlie
 rs or anomalies. In this talk\, I will start with an approach to outlier-r
 esilient principal component analysis\, which establishes a neat link betw
 een the seemingly unrelated notions of sparsity and robustness to outliers
 \, even when the signals involved are not sparse. I will argue that contro
 lling sparsity of model residuals leads to statistical learning algorithms
  that are computationally affordable and universally robust. The impact of
  these ideas will be demonstrated in applications as diverse as identifica
 tion of aberrant responses in personality assessment surveys\, and unveili
 ng communities in social networks\, as well as intruders from video survei
 llance data. In the second part of the talk\, I will switch focus towards 
 the important task of unveiling and mapping-out network anomalies given li
 nk-level traffic measurements. Leveraging the low intrinsic-dimensionality
  of end-to-end network flows and the sparse nature of anomalies\, I will s
 how how to construct an estimated map of anomalies in real time to aid in 
 monitoring the network health state. If time allows\, I will finally highl
 ight additional application domains that include predicting network-wide p
 ath latencies\, and load curve cleansing and imputation -- a critical task
  in green grid analytics and energy management with renewables.\n\n\nBiogr
 aphy\nG. B.  Giannakis (IEEE Fellow'97) received his Diploma in Electrica
 l Engr.  from the Ntl. Tech. Univ. of Athens\, Greece\, 1981. From 1982 t
 o 1986 he was with the Univ. of Southern California (USC)\, where he recei
 ved his MSc. in Electrical Engineering\, 1983\, MSc. in Mathematics\, 1986
 \,\nand Ph.D. in Electrical Engr.\, 1986. From 1987 to 1998 he was with th
 e Univ. of Virginia. Since 1999 he has been a professor with the Univ. of 
 Minnesota\, where he now holds an ADC Chair in Wireless Telecommunications
  in the ECE Department\, and serves as director of the Digital Technology 
 Center.\n\nHis general interests span the areas of communications\, networ
 king and statistical signal processing - subjects on which he has publishe
 d more than 340 journal papers\, 560 conference papers\, 20 book chapters\
 , two edited books and two research monographs (h-index 99). Current resea
 rch focuses on compressive sensing\, cognitive radios\, cross-layer design
 s\, wireless sensors\, social and power grid networks. He is the (co-)inve
 ntor of 21 patents issued\, and the (co-) recipient of 8 best paper awards
  from the IEEE Signal Processing (SP) and Communications Societies\, inclu
 ding the G. Marconi Prize Paper Award in Wireless Communications. He also 
 received Technical Achievement Awards from the SP Society (2000)\, from EU
 RASIP (2005)\, a Young Faculty Teaching Award\, and the G. W. Taylor Award
  for Distinguished Research from the University of Minnesota. He is a Fell
 ow of EURASIP\, the IEEE (1997)\, and has served the IEEE in a number of p
 osts\, including that of a Distinguished Lecturer for the IEEE-SP Society.
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR