Empirical metal-oxide RRAM device endurance and retention model for deep learning simulations

Abstract

Memristive devices including resistive random access memory (RRAM) cells are promising nanoscale low-power components projected to facilitate significant improvement in power and speed of Deep Learning (DL) accelerators, if structured in crossbar architectures. However, these devices possess non-ideal endurance and retention properties, which should be modeled efficiently. In this paper, we propose a novel generalized empirical metal-oxide RRAM endurance and retention model for use in large-scale DL simulations. To the best of our knowledge, the proposed model is the first to unify retention-endurance modeling while taking into account time, energy, SET-RESET cycles, device size, and temperature. We compare the model to state-of-the-art and demonstrate its versatility by applying it to experimental data from fabricated devices. Furthermore, we use the model for CIFAR-10 dataset classification using a large-scale deep memristive neural network (DMNN) implementing the MobileNetV2 architecture. Our results show that, even when ignoring other device non-idealities, retention and endurance losses significantly affect the performance of DL networks. Our proposed model and its DL simulations are made publicly available.

Effect of Post-Annealing on Barrier Modulations in Pd/IGZO/SiO2/p+-Si Memristors

“…It is essential to develop a high-performance non-volatile memory that can be implemented as hardware. The performance indicators of non-volatile memory are linearity [ 10 , 11 , 12 , 13 ], retention [ 14 , 15 , 16 ], endurance [ 17 , 18 , 19 ], the number of conductance states [ 20 , 21 , 22 , 23 ], power consumption, [ 24 , 25 , 26 , 27 ] and device variation [ 28 , 29 , 30 , 31 , 32 ]. Improvement of synaptic characteristics by optimization of the device can simply enhance the performance of neuromorphic systems.…”

Resistive Switching Characteristics of ZnO-Based RRAM on Silicon Substrate

“…Recently, optimum gate voltage for improving endurance while maintaining a high resistance ratio and the insensitivity of data retention of the HRS state to temperature and cycling-aging was studied [16]. Additionally, the endurance failure in a scaled RRAM device through the vacancy mobility gradation was studied, and the endurance and retention model of RRAM devices for deep learning simulations has also been studied [17].…”

An Interface‐Type Memristive Device for Artificial Synapse and Neuromorphic Computing

“…Such a device is not only capable of storing information, but it can also emulate biological synaptic functions, and thus it can serve as a fundamental hardware element in the neuromorphic computing. [8,11] In a biological synapse, two neurons (presynaptic and postsynaptic neurons) are connected by a synapse, and the analog reconfiguration of synaptic plasticity (connection strength) due to synaptic spikes is key to learning and memory in a human brain. [12][13][14] Similarly, a two-terminal memristor can structurally and functionally mimic biological synaptic features such as the conductance (G), corresponding to a synaptic weight, which is updated based on the history of its input potential.…”