Constant memory implementation of a phase-model neural network
Disclosed are systems, apparatuses, and methods for implementing a phase-model neural network using a fixed amount of memory. Such a phase-model neural network includes a plurality of neurons, wherein each neuron is associated with two parameters—an activity and a phase. Example methods include (i) generating a sequence of variables associated with a probability distribution of phases and (ii) sequentially sampling the probability distribution of phases using a fixed amount of memory, regardless of a number of phases used in the phase-model neural network.
1. A method for implementing a phase-model neural network, the phase-model neural network comprising a plurality of neurons, with individual ones of the plurality of neurons being associated with an activity level and a phase of firing, the method comprising:
obtaining a probability distribution of phases;
normalizing the probability distribution based on a predetermined number of phases used in the phase-model neural network; and
sequentially sampling the normalized probability distribution of phases, based on a comparison with a threshold value and either a random number and/or a pseudo-random number, using no more than a fixed amount of memory to store the probability distribution of phases, regardless of the predetermined number of phases used in the phase-model neural network;
updating individual ones of the plurality of neurons based on the sequentially sampled phases.