Neuromorphic Algorithm-hardware Codesign for Temporal Pattern Learning

Fang, Haowen; Taylor, Brady; Li, Ziru; Mei, Zaidao; Li, Hai; Qiu, Qinru

doi:10.1109/dac18074.2021.9586133

Cited by 12 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the neural dynamics, neuroscientists have developed many neural models, represented by the Hodgkin-Huxley (Hodgkin et al, 1952 ), the leaky integrate-and-fire (LIF) (Dayan and Abbott, 2005 ), and the Izhikevich (Izhikevich et al, 2004 ) model. As a simplified model of neuron dynamics, LIF has garnered a great deal of interest from algorithm researchers and has been extensively implemented in SNNs (Lee et al, 2016 ; Wu et al, 2018 ; Cheng et al, 2020 ; Yin et al, 2020 ; Fang et al, 2021 ). The LIF model captures the intuitive properties of external input accumulating charge across a leaky neural membrane with a clear threshold (Tavanaei et al, 2019 ).…”

Section: Methodsmentioning

confidence: 99%

“…In SNNs, synaptic strengths are described as scalar weights that can be dynamically modified according to a particular learning rule. Actively investigated, the learning rules of SNNs can be generally categorized into three categories: conversion-based methods that map SNNs from trained ANNs (Diehl et al, 2016 ; Hunsberger and Eliasmith, 2016 ; Rueckauer et al, 2016 , 2017 ; Sengupta et al, 2019 ; Han et al, 2020 ); supervised learning with spikes that directly train SNNs using variations of error backpropagation (Lee et al, 2016 ; Shrestha and Orchard, 2018 ; Wu et al, 2018 , 2019 ; Neftci et al, 2019 ; Yin et al, 2020 ; Fang et al, 2021 ); local learning rules at synapses, such as schemes exploring the spike time dependent plasticity (STDP) (Song et al, 2000 ; Nessler et al, 2009 ; Diehl and Cook, 2015 ; Tavanaei et al, 2016 ; Masquelier and Kheradpisheh, 2018 ). In addition to the above-mentioned directions, many new algorithms have emerged, such as: a biological plausible BP implementation in pyramidal neurons based on the Bursting mechanism (Payeur et al, 2021 ); a biologically plausible online learning based on rewards and eligibility traces (Bellec et al, 2020 ); and the target-based learning in recurrent spiking networks (Ingrosso and Abbott, 2019 ; Muratore et al, 2021 ), which provides an alternative to error-based approaches.…”

Section: Methodsmentioning

confidence: 99%

“…It deserves to mention that Wu et al ( 2018 ) presented the Spatio-Temporal Backpropagation (STBP) technique, which for the first time translated the discretized Leaky-Integrate-and-Fire (LIF) neural model into the Pytorch framework (Paszke et al, 2019 ), significantly advancing the advancement of direct training approaches. Since then, several research works on SNN algorithms based on the Backpropagation-Through-Time (BPTT) algorithms and the LIF neural model have appeared, consistently showing the capability and performance of SNN (Lee et al, 2016 ; Shrestha and Orchard, 2018 ; Wu et al, 2018 , 2019 ; Neftci et al, 2019 ; Yin et al, 2020 ; Fang et al, 2021 ). We refer to these direct training techniques as BP-based SNN, or BP-SNN for short.…”

Section: Introductionmentioning

confidence: 99%

“…Initially, BPTT-based SNNs needed a high number of time steps to verify that discrete models and LIF neuron dynamics corresponded (Wu et al, 2018 ). However, recent work has reduced the number of time steps to make inference computation faster and more efficient, while also lowering SNN calculation latency (Wu et al, 2019 ; Yin et al, 2020 ; Chowdhury et al, 2021 ; Fang et al, 2021 ). When total time is constant, a lesser number of time steps corresponds to a larger step length (Δ t ), which may violate the pre-restriction (Δ t → 0) of the discrete LIF model.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MAP-SNN: Mapping spike activities with multiplicity, adaptability, and plasticity into bio-plausible spiking neural networks

Chen

et al. 2022

Front. Neurosci.

View full text Add to dashboard Cite

Spiking Neural Networks (SNNs) are considered more biologically realistic and power-efficient as they imitate the fundamental mechanism of the human brain. Backpropagation (BP) based SNN learning algorithms that utilize deep learning frameworks have achieved good performance. However, those BP-based algorithms partially ignore bio-interpretability. In modeling spike activity for biological plausible BP-based SNNs, we examine three properties: multiplicity, adaptability, and plasticity (MAP). Regarding multiplicity, we propose a Multiple-Spike Pattern (MSP) with multiple-spike transmission to improve model robustness in discrete time iterations. To realize adaptability, we adopt Spike Frequency Adaption (SFA) under MSP to reduce spike activities for enhanced efficiency. For plasticity, we propose a trainable state-free synapse that models spike response current to increase the diversity of spiking neurons for temporal feature extraction. The proposed SNN model achieves competitive performances on the N-MNIST and SHD neuromorphic datasets. In addition, experimental results demonstrate that the proposed three aspects are significant to iterative robustness, spike efficiency, and the capacity to extract spikes' temporal features. In summary, this study presents a realistic approach for bio-inspired spike activity with MAP, presenting a novel neuromorphic perspective for incorporating biological properties into spiking neural networks.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MAP-SNN: Mapping spike activities with multiplicity, adaptability, and plasticity into bio-plausible spiking neural networks

Chen

et al. 2022

Front. Neurosci.

View full text Add to dashboard Cite

show abstract

“…In [249], Fang et al propose neuromorphic learning and hardware co-design for temporal pattern learning. Authors first propose an efficient training algorithm for SNN with LIF neurons to learn from temporal data.…”

Section: Hardware-software Co-designmentioning

confidence: 99%

Implementing Spiking Neural Networks on Neuromorphic Architectures: A Review

Huynh¹,

Varshika²,

Paul³

et al. 2022

Preprint

View full text Add to dashboard Cite

Recently, both industry and academia have proposed several different neuromorphic systems to execute machine learning applications that are designed using Spiking Neural Networks (SNNs). With the growing complexity on design and technology fronts, programming such systems to admit and execute a machine learning application is becoming increasingly challenging. Additionally, neuromorphic systems are required to guarantee real-time performance, consume lower energy, and provide tolerance to logic and memory failures. Consequently, there is a clear need for system software frameworks that can implement machine learning applications on current and emerging neuromorphic systems, and simultaneously address performance, energy, and reliability. Here, we provide a comprehensive overview of such frameworks proposed for both, platform-based design and hardware-software co-design. We highlight challenges and opportunities that the future holds in the area of system software technology for neuromorphic computing.

show abstract

RN‐Net: Reservoir Nodes‐Enabled Neuromorphic Vision Sensing Network

Yoo,

Lee,

Wang

et al. 2024

Advanced Intelligent Systems

View full text Add to dashboard Cite

Neuromorphic computing systems promise high energy efficiency and low latency. In particular, when integrated with neuromorphic sensors, they can be used to produce intelligent systems for a broad range of applications. An event‐based camera is such a neuromorphic sensor, inspired by the sparse and asynchronous spike representation of the biological visual system. However, processing the event data requires either using expensive feature descriptors to transform spikes into frames, or using spiking neural networks (SNNs) that are expensive to train. In this work, a neural network architecture is proposed, reservoir nodes‐enabled neuromorphic vision sensing network (RN‐Net), based on dynamic temporal encoding by on‐sensor reservoirs and simple deep neural network (DNN) blocks. The reservoir nodes enable efficient temporal processing of asynchronous events by leveraging the native dynamics of the node devices, while the DNN blocks enable spatial feature processing. Combining these blocks in a hierarchical structure, the RN‐Net offers efficient processing for both local and global spatiotemporal features. RN‐Net executes dynamic vision tasks created by event‐based cameras at the highest accuracy reported to date at one order of magnitude smaller network size. The use of simple DNN and standard backpropagation‐based training rules further reduces implementation and training costs.

show abstract

Neuromorphic Algorithm-hardware Codesign for Temporal Pattern Learning

Cited by 12 publications

References 15 publications

MAP-SNN: Mapping spike activities with multiplicity, adaptability, and plasticity into bio-plausible spiking neural networks

MAP-SNN: Mapping spike activities with multiplicity, adaptability, and plasticity into bio-plausible spiking neural networks

Implementing Spiking Neural Networks on Neuromorphic Architectures: A Review

RN‐Net: Reservoir Nodes‐Enabled Neuromorphic Vision Sensing Network

Contact Info

Product

Resources

About