Bio-inspired computing with resistive memories &amp;#x2014; models, architectures and applications

Wu, Qing; Liu, Beiye; Chen, Yiran; Li, Hai; Chen, Qiuwen; Qiu, Qinru

doi:10.1109/iscas.2014.6865265

Cited by 10 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first neuron entity (N1), and the associated tripartite synapses, spike and probability generators, are mapped into node (0,1) in the NoC mesh. N2 and its associated blocks are mapped to node (1,0) while the astrocyte core is mapped to node (0,0) and FCMP is located at node (1,1). FCMP sends configuration packets at the start of a simulation, routing information from the SANN components to their destination.…”

Section: B Multi-fpga Sann Implementationmentioning

confidence: 99%

See 1 more Smart Citation

AstroByte: Multi-FPGA Architecture for Accelerated Simulations of Spiking Astrocyte Neural Networks

Karim

Harkin

McDaid

et al. 2020

2020 Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE)

View full text Add to dashboard Cite

Spiking astrocyte neural networks (SANN) are a new computational paradigm that exhibit enhanced self-adapting and reliability properties. The inclusion of astrocyte behaviour increases the computational load and critically the number of connections, where each astrocyte typically communicates with 6-9 neurons (and their associated synapses) with feedback pathways from each neuron to the astrocyte. Each astrocyte cell also communicates with its neighbouring cell resulting in a significant interconnect density. The substantial level of parallelisms in SANNs lends itself to acceleration in hardware, however, the challenge in accelerating simulations of SANNs firmly resides in scalable interconnect and the ability to inject and retrieve data from the hardware. This paper presents a novel multi-FPGA acceleration architecture, AstroByte, for the speedup of SANNs. AstroByte explores Networks-on-Chip (NoC) routing mechanisms to address the challenge of communicating both spike event (neuron data) and numeric (astrocyte data) across significant interconnect pathways between astrocytes and neurons. AstroByte also exploits the NoC interconnect to inject data and retrieve runtime data from the accelerated SANN simulations. Results will show that AstroByte can simulate SANN applications with speedup factors of between x162-x188 over Matlab equivalent simulations.

show abstract

Section: B Multi-fpga Sann Implementationmentioning

confidence: 99%

“…The human brain can carry out computations in a power efficient and immensely parallel manner which has motivated the trend in bio-inspired computing [1]. Spiking Neural Networks (SNNs) are a popular bio-inspired paradigm that have been used in many applications [2].…”

Section: Introductionmentioning

confidence: 99%

AstroByte: Multi-FPGA Architecture for Accelerated Simulations of Spiking Astrocyte Neural Networks

Karim

Harkin

McDaid

et al. 2020

2020 Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE)

View full text Add to dashboard Cite

show abstract

“…A memristance (the memristor resistance) ( Figure 1) can be represented as (cf. [10]) where 0 ≤ p ≤ 1 is the doping front position relative to the total film thickness h of TiO 2 , R on is the memristor minimum resistance, R off is the memristor maximum resistance. When a voltage V above a certain threshold V th is applied to the memristor, its memristance decreases due to the expansion of the doped band D having a low resistance and reducing the zone U of pure oxide having a high resistance.…”

Section: Memristor Crossbarmentioning

confidence: 99%

The Memristor Crossbar-Based Wta Neural Network

Tarkov¹,

Osipov²

2016

bncc.cs

View full text Add to dashboard Cite

The problems of programming memristor arrays (memristor crossbars) are considered. An estimate for the pulse width to set the desired memristor resistance (memristance) value is obtained. The implementation of the Winner-Take-All (WTA) neural network on the memristor crossbar and the NMOS transistors for binary images recognition is proposed. The proposed WTA network implementation by simulation on the LTspice IV software was approved.

show abstract

“…The human brain can carry out computations in a powerefficient and massively parallel manner which has motivated the trend in Bio-inspired computing [1]. Spiking Neural Networks (SNNs) are a popular bio-inspired paradigm that have been used in many applications [2].…”

Section: Introductionmentioning

confidence: 99%

FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware

Karim

Harkin

McDaid

et al. 2018

2018 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

Spiking Astrocyte-neuron Networks (SANNs) model the adaptive/repair feature of the human brain. They integrate astrocyte cells with spiking neurons to facilitate a distributed and fine-grained self-repair capability at the synapse level. SANNs are more complex with the addition of astrocyte cells and require longer simulation times, as they are dynamic over much longer timescales than traditional neural networks. Therefore, dedicated FPGA accelerators offer reductions in simulation times. To support the acceleration of SANNs, the capability of fault injection to synapses and monitoring significant levels of neuron and astrocyte data for off-chip transmission to PC-based analysis, are required. This paper presents an FPGA-based monitoring platform (FMP) for injecting faults and capturing and analyzing data acquired from the SANN FPGA accelerator, Astrobyte. The FMP uses custom logic and a NIOS II based system to control fault injection and data monitoring on the FPGA. Results show accurate accelerated simulations of fault injection scenarios using FMP with speedups up to 65 times greater compared with equivalent Matlab implementations.

show abstract

Bio-inspired computing with resistive memories — models, architectures and applications

Cited by 10 publications

References 21 publications

AstroByte: Multi-FPGA Architecture for Accelerated Simulations of Spiking Astrocyte Neural Networks

AstroByte: Multi-FPGA Architecture for Accelerated Simulations of Spiking Astrocyte Neural Networks

The Memristor Crossbar-Based Wta Neural Network

FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware

Contact Info

Product

Resources

About