Wh. Wong et Fm. Liang, DYNAMIC WEIGHTING IN MONTE-CARLO AND OPTIMIZATION, Proceedings of the National Academy of Sciences of the United Statesof America, 94(26), 1997, pp. 14220-14224
Dynamic importance weighting is proposed as a Monte Carlo method that
has the capability to sample relevant parts of the configuration space
even in the presence of many steep energy minima. The method relies o
n an additional dynamic variable (the importance weight) to help the s
ystem overcome steep barriers. A non-Metropolis theory is developed fo
r the construction of such weighted samplers. Algorithms based on this
method are designed for simulation and global optimization tasks aris
ing from multimodal sampling, neural network training, and the traveli
ng salesman problem. Numerical tests on these problems confirm the eff
ectiveness of the method.