Abstract
For realizing neural networks with binary memristor crossbars, memristors should be programmed by high-resistance state (HRS) and low-resistance state (LRS), according to the training algorithms like backpropagation. Unfortunately, it takes a very long time and consumes a large amount of power in training the memristor crossbar, because the program-verify scheme of memristor-programming is based on the incremental programming pulses, where many programming and verifying pulses are repeated until the target conductance. Thus, this reduces the programming time and power is very essential for energy-efficient and fast training of memristor networks. In this paper, we compared four different programming schemes, which are F-F, C-F, F-C, and C-C, respectively. C-C means both HRS and LRS are coarse-programmed. C-F has the coarse-programmed HRS and fine LRS, respectively. F-C is vice versa of C-F. In F-F, both HRS and LRS are fine-programmed. Comparing the error-energy products among the four schemes, C-F shows the minimum error with the minimum energy consumption. The asymmetrical coarse HRS and fine LRS can reduce the time and energy during the crossbar training significantly, because only LRS is fine-programmed. Moreover, the asymmetrical C-F can maintain the network’s error as small as F-F, which is due to the coarse-programmed HRS that slightly degrades the error.
Highlights
The increase in the number of edge devices such as mobile devices, smart phones, etc. demand a larger amount of data processing with higher energy efficiency
Since many memristors are based on the filamentary conduction mechanism with the abrupt transition between low-resistance state (LRS) and high-resistance state (HRS) [26], in this paper, we focus on binary memristor crossbars
For realizing neural networks with binary memristor crossbars, memristors should be programmed by HRS and LRS, according to the training algorithms such as the backpropagation [30]
Summary
The increase in the number of edge devices such as mobile devices, smart phones, etc. demand a larger amount of data processing with higher energy efficiency. Edge computing is defined as the method that can move the control of data processing from the high-performance cloud systems to the last-edge devices of Internet of Things such as smart sensors, etc., where various data are sensed, collected, and generated from the physical environment. In order to interpret a vast amount of unstructured data from the physical world, neural network techniques such as deep learning should be embedded in the edge-computing devices. It is widely expected that the edge computing technique may be in the mainstream in 2 to 5 years, as neural-network-based deep learning, Internet of Things, and smart sensors contribute to each other mutually to advance the edge-computing technique further [3,4]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
