It is summer in an even year, which means it is time for the Eleventh World Meeting of ISBA, this time held in Kyoto, Japan. For the first time I am organizing a session here, and we have some great speakers lined up: Galin Jones, James Flegal, Alex Beskos, and yours truly. My talk is: Fast approximation algorithms for partition functions of Gibbs distributions, and has an abstract very close to my longer talk at Ohio State last month.
Abstract: A new algorithm for estimating the partition function of a Gibbs distribution is presented. The method combines a well balanced cooling schedule created through TPA and a product importance sampler. One advantage of the algorithm over existing methods is the standard deviation of the estimate can be bounded theoretically. The number of samples necessary to build a close estimate grows almost linearly in the logarithm of the partition function, making the approach suitable for high dimensional problems. The samples needed for the estimate can be generated rapidly by methods such as parallel tempering.