BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Adaptive Sensing and Information - Prof Rob Nowak\, University of 
 Wisconsin-Madison
DTSTART:20100527T131500Z
DTEND:20100527T141500Z
UID:TALK24841@talks.cam.ac.uk
CONTACT:Rachel Fogg
DESCRIPTION:The engineering and scientific study of large-scale systems is
  now a major focus in technology\, biology\, sociology\, and cognitive sci
 ence. Deciding where\, when\, what and how to `sense' or `measure' is a cr
 ucial question in the scientific study of such systems. For example\, to u
 nderstand the nature of the Internet or an ecosystem one would ideally lik
 e to monitor it everywhere and all the time. This is impossible\, and cons
 equently\, engineers and scientists must select a relatively small number 
 of times\, places and features for measurement.\nSimilar problems arise in
  systems biology. Biological systems are not defined by the independent fu
 nctions of individual genes\, but rather they depend on the complex intera
 ctions of thousands of genes\, proteins\, and small molecules.  The dynami
 c interplay of these elements is precisely coordinated by signaling networ
 ks that orchestrate their interactions. High-throughput experimental techn
 iques now provide biologists with incredibly rich and potentially revealin
 g datasets. But it is impossible to exhaustively explore the full experime
 ntal space\, and so scientists must judiciously choose which experiments t
 o perform. The complexity and high-dimensionality of these systems makes i
 t incredibly difficult for humans alone to manage and optimize sensing pro
 cesses. A grand challenge for engineering and scientific discovery in the 
 21st century is to devise machines that automatically control and adapt se
 nsing operations in large-scale systems.\n\nTraditional approaches to sens
 ing and information processing are non-adaptive in the sense that all data
  are collected prior to analysis and processing. One can envision\, howeve
 r\, adaptive strategies in which information gleaned from previously colle
 cted data is used to guide the selection of new data. This talk focuses on
  the emerging theory and practice of adaptive sensing systems. I will show
  that automatic feedback from data analysis to data collection can be cruc
 ial for effective learning and inference.  To illustrate the role of feedb
 ack\, I will describe two adaptive sensing systems. First\, I will conside
 r binary-valued prediction (classification) problems arising\, for example
 \, in wireless sensing applications.  Under a fixed sensing or communicati
 on budget\, the prediction errors of adaptive sensor networks can be expon
 entially smaller than those of non-adaptive sensing systems. Second\, moti
 vated by problems in systems biology\, I will discuss the role of adaptive
  sensing in the recovery of sparse signals in noise. I will show that cert
 ain weak\, sparse patterns are imperceptible in non-adaptive measurements\
 , but can be recovered perfectly using adaptive sensing and experimental d
 esign.\n
LOCATION:LR12\, Engineering\, Department of
END:VEVENT
END:VCALENDAR