Neuromorphic computing with multi-memristive synapses

Abstract

Neuromorphic computing has emerged as a promising avenue towards building the next generation of intelligent computing systems. It has been proposed that memristive devices, which exhibit history-dependent conductivity modulation, could efficiently represent the synaptic weights in artificial neural networks. However, precise modulation of the device conductance over a wide dynamic range, necessary to maintain high network accuracy, is proving to be challenging. To address this, we present a multi-memristive synaptic architecture with an efficient global counter-based arbitration scheme. We focus on phase change memory devices, develop a comprehensive model and demonstrate via simulations the effectiveness of the concept for both spiking and non-spiking neural networks. Moreover, we present experimental results involving over a million phase change memory devices for unsupervised learning of temporal correlations using a spiking neural network. The work presents a significant step towards the realization of large-scale and energy-efficient neuromorphic computing systems.

References in corpus (6)

• Training and Operation of an Integrated Neuromorphic Network Based on Metal-Oxide Memristors
• Integration of nanoscale memristor synapses in neuromorphic computing architectures
• Acceleration of Deep Neural Network Training with Resistive Cross-Point Devices
• Mixed-Precision In-Memory Computing
• Temporal correlation detection using computational phase-change memory
• Brain-like associative learning using a nanoscale non-volatile phase change synaptic device array

Cited by in corpus (48)

• Parallel convolution processing using an integrated photonic tensor core
• Accurate deep neural network inference using computational phase-change memory
• All-Optically Controlled Memristor
• A 64-core mixed-signal in-memory compute chip based on phase-change memory for deep neural network inference
• The promise of spintronics for unconventional computing
• Hafnia-based Double Layer Ferroelectric Tunnel Junctions as Artificial Synapses for Neuromorphic Computing
• Mixed-precision deep learning based on computational memory
• Memristive control of mutual SHNO synchronization for neuromorphic computing
• Event-Based Backpropagation can compute Exact Gradients for Spiking Neural Networks
• A back-end, CMOS compatible ferroelectric Field Effect Transistor for synaptic weights
• A recipe for creating ideal hybrid memristive-CMOS neuromorphic computing systems
• A neuromorphic systems approach to in-memory computing with non-ideal memristive devices: From mitigation to exploitation
• On-chip learning for domain wall synapse based Fully Connected Neural Network
• Using the IBM Analog In-Memory Hardware Acceleration Kit for Neural Network Training and Inference
• A memristive deep belief neural network based on silicon synapses
• Scalable Deep Learning on Distributed Infrastructures: Challenges, Techniques and Tools
• A Nanoscale Room-Temperature Multilayer Skyrmionic Synapse for Deep Spiking Neural Networks
• NbOx based memristor as artificial synapse emulating short term plasticity
• Analog Neural Computing with Super-resolution Memristor Crossbars
• Photocurrent Imaging of Multi-Memristive Charge Density Wave Switching in Two-Dimensional 1T-TaS2
• Progressive amorphization of GeSbTe phase-change material under electron beam irradiation
• Minimizing the programming power of phase change memory by using graphene nanoribbon edge-contact
• Energy Efficient Learning with Low Resolution Stochastic Domain Wall Synapse Based Deep Neural Networks
• Noisy Machines: Understanding Noisy Neural Networks and Enhancing Robustness to Analog Hardware Errors Using Distillation
• X-CHANGR: Changing Memristive Crossbar Mapping for Mitigating Line-Resistance Induced Accuracy Degradation in Deep Neural Networks
• Voltage-Dependent Synaptic Plasticity (VDSP): Unsupervised probabilistic Hebbian plasticity rule based on neurons membrane potential
• Stateful Logic using Phase Change Memory
• ETLP: Event-based Three-factor Local Plasticity for online learning with neuromorphic hardware
• 4K-Memristor Analog-Grade Passive Crossbar Circuit
• Spiking Hyperdimensional Network: Neuromorphic Models Integrated with Memory-Inspired Framework
• Online Training of Spiking Recurrent Neural Networks with Phase-Change Memory Synapses
• Spiking neuromorphic chip learns entangled quantum states
• AnalogNets: ML-HW Co-Design of Noise-robust TinyML Models and Always-On Analog Compute-in-Memory Accelerator
• Low-Power Neuromorphic Hardware for Signal Processing Applications
• Phase Transitions in Germanium Telluride Nanoparticle Phase-Change Materials Studied by Time-Resolved X-Ray Diffraction
• Nonuniform magnetic domain-wall synapses enabled by population coding
• Harnessing Optoelectronic Noises in a Photonic Generative Network
• Magnetic-field controlled organic spintronic memristor for neural network computation
• Ultra-low power on-chip learning of speech commands with phase-change memories
• File Classification Based on Spiking Neural Networks
• Noise-Assisted Metastability: From Lévy Flights to Memristors, Quantum Escape, and Josephson-based Axion Searches
• In-situ learning harnessing intrinsic resistive memory variability through Markov Chain Monte Carlo Sampling
• The data center of tomorrow is made up of heterogeneous accelerators
• Artificial Intelligence Accelerators based on Graphene Optoelectronic Devices
• Memristor-Driven Spike Encoding for Fully Implantable Cochlear Implants
• Voltage Controlled Energy Efficient Domain Wall Synapses with Stochastic Distribution of Quantized Weights in the Presence of Thermal Noise and Edge Roughness
• A new step for computing
• 2022 Roadmap on Neuromorphic Computing and Engineering