BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:"Algorithms for Risk-averse Combinatorial Optimization"
DTSTART:20110317T161500
DTSTAMP:20260407T210954Z
UID:74855e7eadef042f5e9576fb89162b901531d3b556747a28cf0c7d75
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Evdokia Nikolova\, MIT\, IC Faculty Candidate\nAbstract:\n
 Optimization has played a key role in making the task of decision\nmaking 
 from art to science in the past century. An important challenge\nthat stil
 l remains is our ability to incorporate the uncertainty in\nour knowledge 
 and risk-aversion in our objective.  A simple but\ninsightful example of t
 his is encapsulated in the decision question:\ngiven a number of route cho
 ices\, which shall I choose?  Interestingly\,\nthis simple question (easil
 y solvable in a deterministic setting)\nbecomes highly non-trivial when we
  incorporate the uncertainty of\ndelays and the individual’s risk-aversi
 on. This primarily stems from\nthe combinatorial nature of the problem cou
 pled with the non-convexity\nof the objective.\n\nIn this talk I explain h
 ow to solve this reliable route planning\nproblem\, and mention how its so
 lution has been adapted in the MIT\nCarTel system for routing\, which inco
 rporates real traffic information\n(cartel.csail.mit.edu).  I then show ho
 w the solution extends to a\ngeneral framework of risk-averse combinatoria
 l optimization\, for which\nI present exact and approximation algorithms. 
  These general-purpose\nalgorithms can also cope with combinatorial proble
 ms that are NP-hard\,\nwhose deterministic versions we only know how to ap
 proximate.  At the\nend\, I touch upon how the risk-averse framework provi
 des a foundation\nfor studying equilibria in stochastic network games.\n\n
 Bio: Evdokia Nikolova is a postdoctoral associate in the Computer Science\
 nand Artificial Intelligence Laboratory at MIT.  She graduated with a\nBA 
 in Applied Mathematics with Economics from Harvard University\, MS\nin Mat
 hematics from the Statistical Laboratory at Cambridge\nUniversity\, U.K. a
 nd Ph.D. in Computer Science from MIT.\nShe is interested in algorithms ar
 ising in stochastic optimization\,\nnetworks and economics.
LOCATION:INM10
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR