Research on Sparsity of Output Synapses in Echo State Networks

Figure 3

Experimental results to verify the existence of redundant output synapses in the small-scale ESN. (a) Curves of the energy efficiency (blue solid line) and the predicted steps (red dotted line) when we silence a synapse each time. The abscissa representing the th synapse is silenced, the left ordinate is the energy efficiency , the right ordinate is the predicted steps, and the horizontal dotted line represents the predicted steps of ESN when all synapses are activated. (b) Curves of energy contribution of 1st, 2nd, 4th, 5^th, and 7th synapses to predictive performance of ESN as time increases. The abscissa is the time, the ordinate represents the energy contribution of synapses to the predictive performance of ESN, and the curves with different colors and different marks represent energy contributions of 1st, 2nd, 4th, 5^th, and 7th synapses to predictive performances of ESN, respectively. (c) Curves of all synaptic weights when the 2nd and 4th synapses are silenced, respectively, and all the synapses are activated. The abscissa represents the th synapse. The ordinate is the value of synaptic weight. Three curves with different colors and different marks represent the curves of all synaptic weights when the 2nd and 4th synapses are silenced, respectively, and all the synapses are activated. (d) Curves of all synaptic weights when the 1st, 5^th, and 7th synapses are silenced, respectively, and all the synapses are activated.

(a)

(b)

(c)

(d)