BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Improving tabu search behavior : approaches via learning and black
 -box optimization
DTSTART:20230710T110000
DTEND:20230710T120000
DTSTAMP:20260414T222636Z
UID:7d318aae6c75b1c651a9104f7c1eb3c9c54b85d899c80c32ec7f0e1d
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Nadia Lahrichi\, Polytechnique Montréal\nMany approache
 s are used to handle uncertainty in stochastic combinatorial optimization 
 problems. In this talk\, we describe the application of a tabu search appr
 oach in a stochastic environment together with a real application in physi
 cian scheduling in a radiotherapy center. The goal is to determine a weekl
 y cyclic schedule that improves the patient flow and shortens the pretreat
 ment duration. High uncertainty is associated with the arrival day\, profi
 le and type of cancer of each patient. Additionally\, two approaches to im
 prove the efficiency of the method are introduced\, both are based on leve
 raging methods that originate outside the field of metaheuristics. The fir
 st one discusses hyperparameters tuning. Research shows that it is a nontr
 ivial task and efficient methods are required to obtain the best possible 
 results. We present how blackbox optimization can help choose the tabu sea
 rch parameters efficiently. We are solving this problem through a Mesh Ada
 ptive Direct Search (MADS) algorithm with no derivative information. The s
 econd one presents a learning algorithm for improving tabu search by reduc
 ing its search space and evaluation effort. The learning tabu search algor
 ithm uses classification methods in order to better motivate moves through
  the search space.
LOCATION:GC G1 515 https://plan.epfl.ch/?room==GC%20G1%20515
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR