Sparsely encoded associative memory: static synaptic noise and static threshold noise

In the present paper, an associative memory model with sparse coding is analyzed by means of the self-consistent signal-to-noise analysis (SCSNA). We discuss some effects of sparseness and the shape of the response function on memory capacity, considering a case using monotonic neurons. The memory capacity strongly depends on the shape of the response function, as well as sparseness. Moreover, a model with static synaptic noise and static noise in the threshold is discussed.