Pulse event decision-making device and method, chip and electronic equipment

Abstract

The invention relates to a pulse event decision device and method, a chip and electronic equipment. In order to solve the problem of inherent time delay of output pulse events of a neural mimicry chip during decision reading, a first counting module is adopted and is configured to count the number of input pulse events of a pulse neural network; a second counting module is configured to count the number of pulse events respectively issued by a plurality of neurons in the pulse neural network output layer; and a decision module is configured to make a decision result according to the counts of the pulse events respectively issued by the plurality of neurons in the pulse neural network output layer when the count of the first counting module reaches a first preset value or the total count of the second counting module reaches a second preset value. Based on technical means such as counting and sliding counting windows, the problem of inherent time delay or insufficient decision data during decision making is solved, and the technical effect of adaptively making a final decision according to the event occurrence speed is achieved.

Description

technical field [0001] The present invention relates to a pulse event decision-making device, method, chip and electronic equipment, in particular to a pulse event decision-making device, method, chip and electronic equipment for adaptive decision-making at the edge of the Internet of Things (IoT) according to the speed of event occurrence. Background technique [0002] Neuromorphic Computing (Neuromorphic Computing) is a new type of computing architecture developed recently. This computing method uses bionic (eye, brain, etc.) methods to complete reasoning on input signals, breaking through the von Neumann architecture for the first time. Event imaging devices, including but not limited to dynamic vision sensors, generate pulse events according to the light and dark changes at the corresponding positions of the pixels, and are in a silent state when there is no change in light and dark, and do not generate redundant data. Frame-based image processing logic is completely dif...

AI Technical Summary

Problems solved by technology

[0005] However, there are still some unreasonable inherent delays in the prior art ( figure 1 ), eliminating these delays will help the chip / system to obtain decision results faster, and win more precious time for the subsequent system to make a more timely response

In addition, it may be because of this inherent delay that in some cases the chip's inference results may be unreliable

Images