TY - GEN
T1 - Speeding up evolution through learning
T2 - 9th Intelligent Information Systems Symposium, IIS'2000
AU - Michalski, Ryszard S.
AU - Cervone, Guido
AU - Kaufman, Kenneth
PY - 2000/12/1
Y1 - 2000/12/1
N2 - This paper reports briefly on the development of a new approach to evolutionary computation, called the Learnable Evolution Model or LEM. In contrast to conventional Darwinian-type evolutionary algorithms that employ mutation and/or recombination, LEM employs machine learning to generate new populations. At each step of evolution, LEM determines hypotheses explaining why certain individuals in the population are superior to others in performing the designated class of tasks. These hypotheses are then instantiated to create a next generation. In the testing studies described here, we compared a program implementing LEM with selected evolutionary computation algorithms on a range optimization problems and a filter design problem. In these studies, LEM significantly outperformed the evolutionary computation algorithms, sometimes speeding up the evolution by two or more orders of magnitude in the number of evolutionary steps (births). LEM was also applied to a real-world problem of designing optimized heat exchangers. The resulting designs matched or - outperformed the best human designs.
AB - This paper reports briefly on the development of a new approach to evolutionary computation, called the Learnable Evolution Model or LEM. In contrast to conventional Darwinian-type evolutionary algorithms that employ mutation and/or recombination, LEM employs machine learning to generate new populations. At each step of evolution, LEM determines hypotheses explaining why certain individuals in the population are superior to others in performing the designated class of tasks. These hypotheses are then instantiated to create a next generation. In the testing studies described here, we compared a program implementing LEM with selected evolutionary computation algorithms on a range optimization problems and a filter design problem. In these studies, LEM significantly outperformed the evolutionary computation algorithms, sometimes speeding up the evolution by two or more orders of magnitude in the number of evolutionary steps (births). LEM was also applied to a real-world problem of designing optimized heat exchangers. The resulting designs matched or - outperformed the best human designs.
UR - http://www.scopus.com/inward/record.url?scp=24344442702&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=24344442702&partnerID=8YFLogxK
U2 - 10.1007/978-3-7908-1846-8_22
DO - 10.1007/978-3-7908-1846-8_22
M3 - Conference contribution
AN - SCOPUS:24344442702
SN - 9783790813098
T3 - Advances in Soft Computing
SP - 243
EP - 256
BT - Intelligent Information Systems -Proceedings of the IIS'2000 Symposium
Y2 - 12 June 2000 through 16 June 2000
ER -