Neural network online learning system based on a memristor

Abstract

The invention discloses a neural network online learning system based on a memristor. A pulse coding mode of a K-bit input vector is improved; the coding pulse corresponding to each bit is expanded into 2m pulses; In this way, the total number of needed coded pulses is K * 2m, each bit of weighted summation calculation is actually carried out for 2m times, finally, summation averaging operation iscarried out at the output end, the influence of accidental factors and noise on the calculation result in the calculation process is reduced through the mode, and therefore the calculation precisionis improved. The memristor array is simultaneously used for forward weighted summation calculation and weight small storage in the neural network; Different from offline learning, The weight in the memristor array needs to be updated once every time a signal is input through online learning, the variable quantity of the weight is mapped into the number of pulses, then the pulses are applied for one-time weight write-in operation, the neural network training speed can be increased, the hardware cost can be reduced, and the power consumption of neural network training is reduced.

Description

technical field [0001] The invention belongs to the field of artificial neural network hardware, and more specifically relates to a memristor-based neural network online learning system. Background technique [0002] In order to meet the challenges of traditional CMOS technology-based neural network hardware platforms in terms of area, speed, power consumption, and "Von Neumann bottleneck", researchers hope to use non-volatile memory devices such as memristors to build neural network hardware. Accelerator, thereby greatly improving the performance of the neural network hardware system. Memristors are used to implement neural network hardware accelerators. On the one hand, the analog conductance characteristics of memristors are used to better represent the weights in synapses or neural network algorithms; on the other hand, cross-arrays based on memristors can achieve parallel The matrix-vector multiplication operation and weight update of . [0003] At present, the resear...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the defects of the prior art, the purpose of the present invention is to solve the technical problem that the memristor-based neural network in the prior art has a slow speed and cannot be applied to online learning

Images