BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:An Evolutionary Approach to Experimental Design for Combinatorial 
 Optimization - Borrotti\, M (Bologna)
DTSTART:20110902T103000Z
DTEND:20110902T110000Z
UID:TALK32627@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:In this presentation we investigate an approach which combines
  statistical methods and optimization algorithms in order to explore a lar
 ge search space when the great number of variables and the economical cons
 traints limit the ability of classical techniques to reach the optimum of 
 a function. The method we propose - the Model Based Ant Colony Design (MAC
 D) - couples real experimentation with simulated experiments and boosts an
  Ant Colony algorithm (Dorigo et al.\, 2004) by means of a simulator (stri
 ctly speaking an emulator)\, i.e. a predictive statistical model. Candidat
 e solutions are generated by computer simulation using Ant Colony Optimiza
 tion\, a probabilistic technique for solving computational problem which c
 onsists in finding good paths through graphs and is based on the foraging 
 behaviour of real ants. The evaluation of the candidate solutions is achie
 ved by physical experiments and is fed back into the simulative phase in a
  recursive way. \n\nThe properties of the proposed approach are studied by
  means of numerical simulations\, testing the algorithm on some mathematic
 al benchmark functions. Generation after generation\, the evolving design 
 requires a small number of experimental points to test\, and consequently 
 a small investment in terms of resources. Furthermore\, since the research
  was inspired by a real problem in Enzyme Engineering and Design\, namely 
 finding a new enzyme with a specific biological function\, we have tested 
 MACD on the real application. The results shows that the algorithm has exp
 lored a region of the sequence space not sampled by natural evolution\, id
 entifying artificial sequences that fold into a tertiary structure closely
  related to the target one. \n\n\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR