TY - GEN
T1 - Mirrored variants of the (1,4)-CMA-ES compared on the noiseless BBOB-2010 testbed
AU - Auger, Anne
AU - Brockhoff, Dimo
AU - Hansen, Nikolaus
PY - 2010/8/30
Y1 - 2010/8/30
N2 - Derandomization by means of mirrored samples has been recently introduced to enhance the performances of (1,λ)-Evolution-Strategies (ESs) with the aim of designing fast and robust stochastic local search algorithms. This paper compares on the BBOB-2010 noiseless benchmark testbed two variants of the (1,4)-CMA-ES where the mirrored samples are used. Independent restarts are conducted up to a total budget of 104D function evaluations, where D is the dimension of the search space. The results show that the improved variants are significantly faster than the baseline (1,4)-CMA-ES on 4 functions in 20D (respectively 7 when using sequential selection in addition) by a factor of up to 3 (on the attractive sector function). In no case, the (1,4)-CMA-ES is significantly faster on any tested target function value in 5D and 20D. Moreover, the algorithm employing both mirroring and sequential selection is significantly better than the algorithm without sequentialism on five functions in 20D with expected running times that are about 20% smaller.
AB - Derandomization by means of mirrored samples has been recently introduced to enhance the performances of (1,λ)-Evolution-Strategies (ESs) with the aim of designing fast and robust stochastic local search algorithms. This paper compares on the BBOB-2010 noiseless benchmark testbed two variants of the (1,4)-CMA-ES where the mirrored samples are used. Independent restarts are conducted up to a total budget of 104D function evaluations, where D is the dimension of the search space. The results show that the improved variants are significantly faster than the baseline (1,4)-CMA-ES on 4 functions in 20D (respectively 7 when using sequential selection in addition) by a factor of up to 3 (on the attractive sector function). In no case, the (1,4)-CMA-ES is significantly faster on any tested target function value in 5D and 20D. Moreover, the algorithm employing both mirroring and sequential selection is significantly better than the algorithm without sequentialism on five functions in 20D with expected running times that are about 20% smaller.
KW - Algorithms
U2 - 10.1145/1830761.1830773
DO - 10.1145/1830761.1830773
M3 - Conference contribution
AN - SCOPUS:77955937494
SN - 9781450300735
T3 - Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10 - Companion Publication
SP - 1559
EP - 1566
BT - Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10 - Companion Publication
T2 - 12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010
Y2 - 7 July 2010 through 11 July 2010
ER -