2000 character limit reached
Joint Training of Deep Boltzmann Machines (1212.2686v1)
Published 12 Dec 2012 in stat.ML and cs.LG
Abstract: We introduce a new method for training deep Boltzmann machines jointly. Prior methods require an initial learning pass that trains the deep Boltzmann machine greedily, one layer at a time, or do not perform well on classifi- cation tasks.