Mapping Transformation Enabled High-Performance and Low-Energy Memristor-Based DNNs

Abstract

When deep neural network (DNN) is extensively utilized for edge AI (Artificial Intelligence), for example, the Internet of things (IoT) and autonomous vehicles, it makes CMOS (Complementary Metal Oxide Semiconductor)-based conventional computers suffer from overly large computing loads. Memristor-based devices are emerging as an option to conduct computing in memory for DNNs to make them faster, much more energy efficient, and accurate. Despite having excellent properties, the memristor-based DNNs are yet to be commercially available because of Stuck-At-Fault (SAF) defects. A Mapping Transformation (MT) method is proposed in this paper to mitigate Stuck-at-Fault (SAF) defects from memristor-based DNNs. First, the weight distribution for the VGG8 model with the CIFAR10 dataset is presented and analyzed. Then, the MT method is used for recovering inference accuracies at 0.1% to 50% SAFs with two typical cases, SA1 (Stuck-At-One): SA0 (Stuck-At-Zero) = 5:1 and 1:5, respectively. The experiment results show that the MT method can recover DNNs to their original inference accuracies (90%) when the ratio of SAFs is smaller than 2.5%. Moreover, even when the SAF is in the extreme condition of 50%, it is still highly efficient to recover the inference accuracy to 80% and 21%. What is more, the MT method acts as a regulator to avoid energy and latency overhead generated by SAFs. Finally, the immunity of the MT Method against non-linearity is investigated, and we conclude that the MT method can benefit accuracy, energy, and latency even with high non-linearity LTP = 4 and LTD = −4.

An overview memristor based hardware accelerators for deep neural network

“…The memristor 30 is a circuit element that acts as a variable resistor, changing its resistance value and controlling the flow of current by acting as a capacitive element during the passage of current. 31 As a novel hardware technology, memristors operate at very small scales and can be utilized for large-scale integrations. 32 When compared to the operational logic of memristors, the classical complementary metal-oxide-semiconductor (CMOS) working method exhibits similar advantages such as small area requirements and low power consumption.…”

“…Indeed, there have been studies conducted on the development of hardware‐based systems using memristor circuit elements. The memristor 30 is a circuit element that acts as a variable resistor, changing its resistance value and controlling the flow of current by acting as a capacitive element during the passage of current 31 . As a novel hardware technology, memristors operate at very small scales and can be utilized for large‐scale integrations 32 .…”