High-Precision Tuning of State for Memristive Devices by Adaptable Variation-Tolerant Algorithm

Abstract

Using memristive properties common for the titanium dioxide thin film devices, we designed a simple write algorithm to tune device conductance at a specific bias point to 1% relative accuracy (which is roughly equivalent to 7-bit precision) within its dynamic range even in the presence of large variations in switching behavior. The high precision state is nonvolatile and the results are likely to be sustained for nanoscale memristive devices because of the inherent filamentary nature of the resistive switching. The proposed functionality of memristive devices is especially attractive for analog computing with low precision data. As one representative example we demonstrate hybrid circuitry consisting of CMOS summing amplifier and two memristive devices to perform analog multiply and accumulate computation, which is a typical bottleneck operation in information processing.

20 pages, 6 figures

References in corpus (1)

• Memory effects in complex materials and nanoscale systems

Cited by in corpus (24)

• Neuro-memristive Circuits for Edge Computing: A review
• Stochastic resonance in a metal-oxide memristive device
• Applicability of Well-Established Memristive Models for Simulations of Resistive Switching Devices
• Applications and Techniques for Fast Machine Learning in Science
• Phenomenological Modeling of Memristive Devices
• Time complexity of in-memory solution of linear systems
• RAELLA: Reforming the Arithmetic for Efficient, Low-Resolution, and Low-Loss Analog PIM: No Retraining Required!
• Endurance Write Speed Tradeoffs in Nonvolatile Memories
• Neuro-Fuzzy Computing System with the Capacity of Implementation on Memristor-Crossbar and Optimization-Free Hardware Training
• Hardware Implementation of Spiking Neural Networks Using Time-To-First-Spike Encoding
• Dynamical attractors of memristors and their networks
• Observation of Highly Nonlinear Resistive Switching of Al2O3/TiO2-x Memristors at Cryogenic Temperature (1.5 K)
• Shock-waves and commutation speed of memristors
• Neural-PIM: Efficient Processing-In-Memory with Neural Approximation of Peripherals
• Long Short-Term Memory Implementation Exploiting Passive RRAM Crossbar Array
• Fast and Accurate Sparse Coding of Visual Stimuli with a Simple, Ultra-Low-Energy Spiking Architecture
• Uncertainty Modeling of Emerging Device-based Computing-in-Memory Neural Accelerators with Application to Neural Architecture Search
• All-in-One Analog AI Hardware: On-Chip Training and Inference with Conductive-Metal-Oxide/HfOx ReRAM Devices
• Temporal State Machines: Using temporal memory to stitch time-based graph computations
• Memristor-based cryogenic programmable DC sources for scalable in-situ quantum-dot control
• Model of the Weak Reset Process in HfOx Resistive Memory for Deep Learning Frameworks
• Efficient Reprogramming of Memristive Crossbars for DNNs: Weight Sorting and Bit Stucking
• Comparison of Update and Genetic Training Algorithms in a Memristor Crossbar Perceptron
• IMSSA: Deploying modern state-space models on memristive in-memory compute hardware