A Weighted Voting Model Of Associative Memory: Experimental Analysis

In a related paper [1], a weighted voting RAM-based associative memory model was proposed, and a theoretical analysis of its performance on binary and random memory sets was given. In this paper, we give an experimental analysis of the weighted voting memory using both binary-random memory sets and more practical memory sets consisting of grayscale face images. The results show that the weighted voting memory offers higher performance over the voting memory on both types of memory sets.