Fully Unsupervised Spike-Rate-Dependent Plasticity Learning With Oxide- Based Memory Devices

Kumar, Manoj; Bezugam, Sai Sukruth; Khan, Sufyan

doi:10.1109/ted.2021.3077346

Cited by 12 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A classification precision of 90.89% is obtained with 10 000 testing samples, which is comparable to the reported state-of-the-art unsupervised SNN simulation results based on synaptic devices (Table ). − The components and updating manner of SNN are illustrated in the Methods section. The confusion matrix of classification is shown in Figure b.…”

Section: Resultsmentioning

confidence: 99%

Multi-Stimuli-Responsive Synapse Based on Vertical van der Waals Heterostructures

Zhou

Tian

et al. 2022

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Brain-inspired intelligent systems demand diverse neuromorphic devices beyond simple functionalities. Merging biomimetic sensing with weight-updating capabilities in artificial synaptic devices represents one of the key research focuses. Here, we report a multiresponsive synapse device that integrates synaptic and optical-sensing functions. The device adopts vertically stacked graphene/h-BN/WSe2 heterostructures, including an ultrahigh-mobility readout layer, a weight-control layer, and a dual-stimuli-responsive layer. The unique structure endows synapse devices with excellent synaptic plasticity, short response time (3 μs), and excellent optical responsivity (105 A/W). To demonstrate the application in neuromorphic computing, handwritten digit recognition was simulated based on an unsupervised spiking neural network (SNN) with a precision of 90.89%, well comparable with the state-of-the-art results. Furthermore, multiterminal neuromorphic devices are demonstrated to mimic dendritic integration and photoswitching logic. Different from other synaptic devices, the research work validates multifunctional integration in synaptic devices, supporting the potential fusion of sensing and self-learning in neuromorphic networks.

show abstract

Section: Resultsmentioning

confidence: 99%

Multi-Stimuli-Responsive Synapse Based on Vertical van der Waals Heterostructures

Zhou

Tian

et al. 2022

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

“…Encoding techniques in SFA. Traditional spike encoders such as Poisson 43 , rate-based encoding 81 , and population encoding 17 , often struggle to capture the complex dynamics inherent to SFA, potentially leading to information loss 24,82 . In the context of biological systems, neural adaptation serves as a crucial tool for calibrating sensitivity across diverse intensity gradients, illumithe need for specialized SFA-based encoders designed to emulate these biological nuances 83 .…”

Section: Challenges and Roadmapmentioning

confidence: 99%

Spike frequency adaptation: bridging neural models and neuromorphic applications

Ganguly,

Bezugam,

Abs

et al. 2024

Commun Eng

Self Cite

View full text Add to dashboard Cite

The human brain’s unparalleled efficiency in executing complex cognitive tasks stems from neurons communicating via short, intermittent bursts or spikes. This has inspired Spiking Neural Networks (SNNs), now incorporating neuron models with spike frequency adaptation (SFA). SFA adjusts these spikes’ frequency based on recent neuronal activity, much like an athlete’s varying sprint speed. SNNs with SFA demonstrate improved computational performance and energy efficiency. This review examines various adaptive neuron models in computational neuroscience, highlighting their relevance in artificial intelligence and hardware integration. It also discusses the challenges and potential of these models in driving the development of energy-efficient neuromorphic systems.

show abstract

“…In the training methods for SNNs, there are mainly two approaches: online and offline learning [14,15]. For online learning, the spike-timing-dependent plasticity or spike-rate-dependent plasticity are adopted as weight-update rules [16,17]. Online learning can offer a robust SNN system by compensating for variations in synaptic devices during real-time learning.…”

Section: Introductionmentioning

confidence: 99%

Simulation and Optimization of IGZO-Based Neuromorphic System for Spiking Neural Networks

Park,

Yun,

Kim

et al. 2024

IEEE J. Electron Devices Soc.

View full text Add to dashboard Cite

In this paper, we conducted a simulation of an indium-gallium-zinc oxide (IGZO)-based neuromorphic system and proposed layer-by-layer membrane capacitor (Cmem) optimization for integrate-and-fire (I&F) neuron circuits to minimize the accuracy drop in spiking neural network (SNN). The fabricated synaptic transistor exhibited linear 32 synaptic weights with a large dynamic range (~846), and an n-type-only IGZO I&F neuron circuit was proposed and verified by HSPICE simulation. The network, consisting of three fully connected layers, was evaluated with an offline learning method employing synaptic transistor and I&F circuit models for three datasets: MNIST, Fashion-MNIST, and CIFAR-10. For offline learning, accuracy drop can occur due to information loss caused by overflow or underflow in neurons, which is largely affected by Cmem. To address this problem, we introduced a layerby-layer Cmem optimization method that adjusts appropriate Cmem for each layer to minimize the information loss. As a result, high SNN accuracy was achieved for MNIST, Fashion-MNIST, and CIFAR-10 at 98.42%, 89.16%, and 48.06%, respectively. Furthermore, the optimized system showed minimal accuracy degradation under device-todevice variation.

show abstract

Fully Unsupervised Spike-Rate-Dependent Plasticity Learning With Oxide- Based Memory Devices

Cited by 12 publications

References 36 publications

Multi-Stimuli-Responsive Synapse Based on Vertical van der Waals Heterostructures

Multi-Stimuli-Responsive Synapse Based on Vertical van der Waals Heterostructures

Spike frequency adaptation: bridging neural models and neuromorphic applications

Simulation and Optimization of IGZO-Based Neuromorphic System for Spiking Neural Networks

Contact Info

Product

Resources

About