Abstract：Restricted Boltzmann machines （RBMs） are often cited as building blocks of a deep belief network （DBN）.By training several RBMs,DBN can be trained quickly to achieve good performance on various machine learning tasks.To further improve the performance of data representation,inspired by sparse coding theory,we propose a novel sparse RBM model in this paper,referred to as AtanRBM.Different from sparse RBM,we encourage the hidden units to be sparse through adding an arctan norm （arctan approximation of L(0) norm） constraint on the probability density space of hidden units directly,rather then constraining the expected activation of every hidden unit to the same low level of sparsity.Some experiments conducted on MNIST dataset show that AtanRBM learns sparser and more discriminative representations compared with the related state-of-the-art models,and then the deep belief network can achieve better classification performance by using layer-wise unsupervised pre-training of two AtanRBMs.
罗剑江,王振友. 一种提高受限玻尔兹曼机性能的反正切函数逼近L(0)范数方法[J]. 小型微型计算机系统, 2016, 37(11): 2562-2566.
LUO Jian-jiang,WANG Zhen-you. Enhancing Performance of Restricted Boltzmann Machine Using Arctan Approximation of L(0) Norm. Journal of Chinese Computer Systems, 2016, 37(11): 2562-2566.