Synaptic circuit, synaptic array and data processing method based on synaptic circuit

Abstract

The embodiments of the present application relate to the field of neural networks. The synaptic circuit provided by the present invention includes: a first memory, a second memory, and a switch assembly; the switch assembly includes a first switch tube, a second switch tube, and a third switch tube; the first end of the first switch tube is connected to the second switch tube. The first end of a memory is connected, the control end of the first switch tube is connected to the first bit line interface; the first end of the second switch tube is connected to the first end of the second memory, the control end of the second switch tube is connected to the first bit line interface The two-bit line interface is connected; the second end of the first memory is connected with the first end of the second memory; the first end of the third switch is connected with the second end of the second memory, and the second end of the third switch is connected with the second end of the second memory. The word line interface is connected, and the third end of the third switching transistor is grounded. Based on the embodiment of the present application, the first memory and the second memory connected in series are respectively controlled by the first switch tube and the second switch tube, so that cross-interference between memories can be reduced during data storage.

Description

technical field [0001] The invention relates to the field of neural networks, in particular to a synaptic circuit, a synaptic array and a data processing method based on the synaptic circuit. Background technique [0002] As the breakthrough cornerstone of contemporary deep learning technology, neural network is a large-scale parallel simple processing unit that can not only store data, but also analyze and make predictions based on the stored data. Many tasks that were considered unimaginable in the past have now been accomplished with the continuous development of neural network technology. Examples include tasks such as image recognition, speech recognition, and finding deep relationships in datasets. [0003] As a structure connecting different neurons in a neural network, a synapse is a medium through which a neuron communicates with other neurons. In the prior art, such as figure 1 The synaptic circuit in which a memristor is connected in series with a switching tub...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a synaptic circuit, a synaptic array and a data processing method based on the synaptic circuit to solve the problems of a switch tube controlling two parallel memristors, complex control methods and high power consumption

Images