FUNCTIONAL ABILITIES OF A STOCHASTIC LOGIC NEURAL NETWORK

We have studied the information processing ability of stochastic logic neural networks, which constitute one of the pulse-coded artificial neural network families. These networks realize pseudoanalog performance with local learning rules using digital circuits, and therefore suit silicon technology. On the other hand, the synaptic weights and the outputs of neurons in stochastic logic are represented by stochastic pulse sequences; therefore these quantities are associated inevitably with nondeterministic coding noise and range limits for the synaptic weights. We have studied the effects of the coding noise and the limited range of the synaptic weights on association ability. We have found analytically that the limited range of the synaptic weights reduces the coding noise and suppresses the degradation of memory storage capacity. To study the effect of the coding noise on an optimization problem, we simulate a probabilistic Hopfield model (Gaussian machine) which has a continuous neuron output function and probabilistic behavior. A proper choice of the coding noise amplitude and scheduling improves the network's solutions of the traveling saleman problem (TSP). These results suggest that stochastic logic may be useful for implementing probabilistic dynamics as well as deterministic dynamics.