Scalable excitatory synaptic circuit design using floating gate based leaky integrators

Kornijcuk, Vladimir; Lim, Hyungkwang; Kim, Inho; Park, Jong‐Keuk; Lee, Wook‐Seong; Choi, Jung-Hae; Choi, Byung Joon; Jeong, Doo Seok

doi:10.1038/s41598-017-17889-8

Cited by 5 publications

(4 citation statements)

References 51 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Larger displacements produce larger ion polarizations and therefore larger currents. Once the ion polarization process reached the spike-like peak, the current declines rapidly in accordance with an exponential decay function with time, i.e., I = I 0 e – t /τ , where I represents the decaying current at time t , I 0 the maximum peak current, and τ the relaxation time. , The relaxation time (τ) may be interpreted as the time necessary for the current to decrease exponentially to 0.368 of its initial value (i.e., 1/e). Smaller τ implies a faster ion relaxation (or rearrangement) within the PEM, which in turn affords a larger ionic conductivity.…”

Section: Results and Discussionmentioning

confidence: 99%

Flexoelectricity in Flexoionic Polymer Electrolyte Membranes: Effect of Thiosiloxane Modification on Poly(ethylene glycol) Diacrylate and Ionic Liquid Electrolyte Composites

Yue

Cao

Lee

et al. 2020

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

The present article entails the generation of flexoelectricity during cantilever bending of a solid polymer electrolyte membrane (PEM), composed of poly(ethylene glycol) diacrylate (PEGDA) precursor and ionic liquid (hexylmethylimidazolium hexafluorophosphate). The effects of thiosiloxane modification of PEGDA precursor on glass transition, ionic conductivity, and flexoelectric performance have been explored as a function of PEM composition. The glass transition temperature (T g) of the PEM declines with increasing thiosiloxane amount in the PEGDA co-network, while the ionic conductivity improves. The PEM/compliant carbonaceous electrodes assemblies were assembled to determine the flexoelectric coefficients by monitoring electrical voltage/current outputs for various PEM compositions under the intermittent square-wave and dynamic oscillatory sine-wave deformation modes. Of particular interest is that the room temperature flexoelectric coefficient exhibits strong frequency dependence in the vicinity of 0.01–10 Hz, suggesting that ion polarization and ion transport through the ion-dipole complexed networks can still be affected by the mobile side chain branches even in the elastic regime of the covalently bonded PEGDA network. The in-depth understanding of the effect of thiosiloxane side chain on flexoelectricity generation is anticipated to have impact on the development of mechanoelectrical energy conversion devices for energy harvesting applications from natural and dynamical environment.

show abstract

Section: Results and Discussionmentioning

confidence: 99%

Flexoelectricity in Flexoionic Polymer Electrolyte Membranes: Effect of Thiosiloxane Modification on Poly(ethylene glycol) Diacrylate and Ionic Liquid Electrolyte Composites

Yue

Cao

Lee

et al. 2020

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

“…This is an easy-toimplement model insomuch that it can readily be implemented in SNN hardware for on-chip learning, for instance, fully digital SNN (e.g. Loihi 5 ) and analog synaptic circuits 68 .…”

Section: Spike Timing-dependent Plasticitymentioning

confidence: 99%

Tutorial: Neuromorphic spiking neural networks for temporal learning

Jeong

2018

Journal of Applied Physics

Self Cite

View full text Add to dashboard Cite

Spiking neural networks (SNN) as time-dependent hypotheses consisting of spiking nodes (neurons) and directed edges (synapses) are believed to offer unique solutions to reward prediction tasks and the related feedback that are classified as reinforcement learning. Generally, temporal difference (TD) learning renders it possible to optimize a model network to predict the delayed reward in an ad hoc manner. Neuromorphic SNNs-networks built using dedicated hardware-particularly leverage such TD learning for not only reward prediction but also temporal sequence prediction in a physical time domain. In this tutorial, such learning in a physical time domain is referred to as temporal learning to distinguish it from conventional TD learning-based methods that generally involve algorithmic (rather than physical) time. This tutorial addresses neuromorphic SNNs for temporal learning from the scratch. It first concerns general characteristics of SNNs including spiking neurons and information coding schemes and then moves on to temporal learning including its general concept, feasible algorithms, and their association with neurophysiological learning rules that have intensively been enriched for the last few decades.Most of neuromorphic systems built using analog building blocks are indeed based on mixed circuit. 1,3,12 The digital neuromorphic prototypes, TrueNorth 2 , SpiNNaker 4 , and Loihi 5 , highlight a fully digital circuit-based strategy in terms of flexibility of network configuration as well as neuron model parameters and learning algorithms. The progress in digital circuit fabrication techniques underpins high-speed and low-power operation of such digital neuromorphic systems. Additionally, a fieldprogrammable gate array (FPGA) is a handy and cost-effective platform for a digital neuromorphic system. To date, various architectures of spiking neurons 13,14 and neuromorphic system architecture for reconfigurable SNN 15 have been built on FPGAs.An emerging strategy is to utilize nonvolatile memory devices, e.g. oxide-based resistive memory, phase-change memory, magnetic tunnel junction, and floating-gate transistor, as synaptic devices. 16,17 Given needs for an enormous number of synapses in a neuromorphic system, replacing even in part mainstream static random access memory or content-addressable memory by such nonvolatile memories can remarkably enhance the areal density of synapses. Additionally, several nonvolatile memories represent multinary states that can further boost the areal density of synapses. 18 Such cutting edge neuromorphic hardware can leverage its capability by the aid of a user-friendly complier, for instance, equipped with graphical user interface (GUI). An example is Nengo 19 , a GUIbased compiler that readily builds an SNN on neuromorphic hardware. Lately, the complier has successfully been applied to Loihi and Braindrop.SNNs largely vary in model complexity. The complexity is generally a measure of fidelity to biological neural network. That is, the more biologically plausible, the more lik...

show abstract

“…The synaptic weights are stored as a charge in a floating gate of a standard MOS transistor. The charge is modified using only the Fowler-Nordheim tunneling mechanism avoiding the significant power dissipation of the commonly used hotcarrier injection [19], [20]. The synaptic weight is changed in an analog way, and the retention time is expected for years.…”

Section: Introductionmentioning

confidence: 99%

Experimental validation of an analog spiking neural network with STDP learning rule in CMOS technology

Polidori

Camisa

Mesri

et al. 2022

2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAI

View full text Add to dashboard Cite

We report the design in CMOS technology and the experimental characterization of an analog spiking neural network with on-chip unsupervised learning. Long-term synaptic memory is implemented using a floating-gate device in a standard 150 nm CMOS process. The neurons are operated with a voltage supply of only 0.4V, allowing an extremely low power dissipation with an energy dissipation per synaptic operation of about 55 fJ. The CMOS chip includes the circuits for implementing real-time learning of the network based on the Spike Time Dependent Plasticity algorithm. During the learning, the neurons produce pulses of ±4.5 V that change the synaptic weight by activating tunneling currents to change the charge in the floating gates.

show abstract

Scalable excitatory synaptic circuit design using floating gate based leaky integrators

Cited by 5 publications

References 51 publications

Flexoelectricity in Flexoionic Polymer Electrolyte Membranes: Effect of Thiosiloxane Modification on Poly(ethylene glycol) Diacrylate and Ionic Liquid Electrolyte Composites

Flexoelectricity in Flexoionic Polymer Electrolyte Membranes: Effect of Thiosiloxane Modification on Poly(ethylene glycol) Diacrylate and Ionic Liquid Electrolyte Composites

Tutorial: Neuromorphic spiking neural networks for temporal learning

Experimental validation of an analog spiking neural network with STDP learning rule in CMOS technology

Contact Info

Product

Resources

About