Introduction to the theory of neural computation
Introduction to the theory of neural computation
Structural learning with forgetting
Neural Networks
Notions of associative memory and sparse coding
Neural Networks - 1996 Special issue: four major hypotheses in neuroscience
Hi-index | 0.00 |
In this paper, we investigate the effect of synaptogenesis on memories in the brain, using the abstract-associative memory model, Hopfield model with the zero-order synaptic decay. Using the numerical simulation, we demonstrate the possibility that synaptogenesis plays a role in maintaining recent memories embedded in the network while avoiding overloading. For the network consisting of 1000 units, it turned out that the minimum decay rate to avoid overloading is 0.02, and the optimal decay rate to maximize the storage capacity is 0.08. We also show that the average numbers of replacement synapses at each learning step corresponding to these two values are 1187 and 21024, respectively.