Energy Efficiency of Machine Learning in Embedded Systems Using Neuromorphic Hardware

Abstract

Recently, the application of machine learning on embedded systems has drawn interest in both the research community and industry because embedded systems located at the edge can produce a faster response and reduce network load. However, software implementation of neural networks on Central Processing Units (CPUs) is considered infeasible in embedded systems due to limited power supply. To accelerate AI processing, the many-core Graphics Processing Unit (GPU) has been a preferred device to the CPU. However, its energy efficiency is not still considered to be good enough for embedded systems. Among other approaches for machine learning on embedded systems, neuromorphic processing chips are expected to be less power-consuming and overcome the memory bottleneck. In this work, we implemented a pedestrian image detection system on an embedded device using a commercially available neuromorphic chip, NM500, which is based on NeuroMem technology. The NM500 processing time and the power consumption were measured as the number of chips was increased from one to seven, and they were compared to those of a multicore CPU system and a GPU-accelerated embedded system. The results show that NM500 is more efficient in terms of energy required to process data for both learning and classification than the GPU-accelerated system or the multicore CPU system. Additionally, limits and possible improvement of the current NM500 are identified based on the experimental results.

Study on the Implementation of a Simple and Effective Memory System for an AI Chip

“…Figure 5 shows the simulation results, indicating that the delay increases as the delay step is increased using the simulation model. In Figure 5, each step can increase the delay with the delay set in #delay in Figure 4: (1) is the result when the delay is made into one step and it pushes about 200 ps, while (2) shows that 6600 ps is pushed because of running the simulation by pushing 33 steps. Among the primitive cells provided by the process, it is possible to use the primitive cells suitable for the desired amount of delay for the testing.…”

“…Initially, it was approached in a mathematical way based on theory, and it was developed and implemented as software [1]. Since the AI technology implemented in software uses the existing CPU, GPU, and memory system, hardware implementation for the AI structure itself was not required [2]. However, as the utilization of AI gradually increased, not only did the implementation of AI through CPU and GPU become necessary, but so did an AI-dedicated process represented by a neural processing unit (NPU) [3].…”

Using Stochastic Computing for Virtual Screening Acceleration

“…Neuromorphic hardware (NH) is a research field which has been propelled by the need of developing high-performance AI systems [20][21][22][23][24]. NH is able to provide timely responses to those applications which require processing huge amounts of data [25].…”

A Configurable and Fully Synthesizable RTL-Based Convolutional Neural Network for Biosensor Applications

“…The neuro-chips are built with multi-electrode arrays; a kind of integrated circuit chip that performs better because of good connectivity and fast and parallel computation. It needs low power and occupies less memory compared to traditional chips [ 15 , 16 , 17 , 18 ]. The research associated with neuro-chips is further classified into three main methods: fully analog, fully digital, and mixed analog/digital methods.…”