ABSTRACT
Based on the understanding of Friston's free energy theory about brain perception and reasoning, this chapter combines free energy theory with restricted Boltzmann machine to establish a visual information memory model and realize the storage and extraction of visual information. Experimental results show that the classification performance of the proposed method is better than that of SOINN and SVM under different number of training samples, and the classification effect is better than IPAMM model when there are more training samples (more than 10).
