Integration of Neuromorphic AI in Event-Driven Distributed Digitized Systems Concepts and Research Directions

2025-05-05 0 0 534.42KB 31 页 10玖币

侵权投诉

Integration of Neuromorphic AI

in Event-Driven Distributed Digitized Systems:

Concepts and Research Directions

Mattias Nilsson∗, Olov Schel´en∗, Anders Lindgren∗†, Ulf Bodin∗,

Cristina Paniagua∗, Jerker Delsing∗, and Fredrik Sandin∗

October 21, 2022

Abstract

Increasing complexity and data-generation rates in cyber-physical sys-

tems and the industrial Internet of things are calling for a corresponding

increase in AI capabilities at the resource-constrained edges of the In-

ternet. Meanwhile, the resource requirements of digital computing and

deep learning are growing exponentially, in an unsustainable manner.

One possible way to bridge this gap is the adoption of resource-eﬃcient

brain-inspired “neuromorphic” processing and sensing devices, which use

event-driven, asynchronous, dynamic neurosynaptic elements with colo-

cated memory for distributed processing and machine learning. However,

since neuromorphic systems are fundamentally diﬀerent from conventional

von Neumann computers and clock-driven sensor systems, several chal-

lenges are posed to large-scale adoption and integration of neuromorphic

devices into the existing distributed digital–computational infrastructure.

Here, we describe the current landscape of neuromorphic computing, fo-

cusing on characteristics that pose integration challenges. Based on this

analysis, we propose a microservice-based framework for neuromorphic

systems integration, consisting of a neuromorphic-system proxy, which

provides virtualization and communication capabilities required in dis-

tributed systems of systems, in combination with a declarative program-

ming approach oﬀering engineering-process abstraction. We also present

concepts that could serve as a basis for the realization of this framework,

and identify directions for further research required to enable large-scale

system integration of neuromorphic devices.

1 Introduction

The accelerating developments of digital computing technology and deep learning–

based AI are leading towards technological, environmental, and economic im-

passes (Thompson et al., 2021; Mehonic and Kenyon, 2022). With the end

of Dennard transistor-scaling (Davari et al., 1995) and the anticipated end

∗Embedded Intelligent Systems Lab (EISLAB), Lule˚a University of Technology, Lule˚a,

Sweden. Email: mattias.1.nilsson@ltu.se

†Applied AI and IoT, RISE Research Institutes of Sweden, Kista, Sweden

arXiv:2210.11190v1 [cs.NE] 20 Oct 2022

of Moore’s law (Waldrop, 2016; Shalf, 2020; Leiserson et al., 2020), conven-

tional digital von Neumann computers and clock-driven sensor systems face

considerable hurdles regarding bandwidth and computational eﬃciency. For

example, the gap between the computational requirements for training state-of-

the-art deep learning models and the capacity of the underlying hardware has

grown exponentially during the last decade (Mehonic and Kenyon, 2022). Mean-

while, in stark contrast, distributed digitized systems—ever-growing in size and

complexity—require increasing computational eﬃciency for AI applications at

the resource-constrained edge of the internet (Zhou et al., 2019; Ye et al., 2021),

where sensors are generating increasingly unmanageable amounts of data.

One approach to addressing this lack of computational capacity and eﬃ-

ciency is oﬀered by neuromorphic engineering (Mead, 1990, 2020). There, in-

spiration is drawn from the most eﬃcient information processing systems known

to humanity—brains—for the design of hardware systems for sensing (Tayarani-

Najaran and Schmuker, 2021) and processing (Zhang et al., 2020a; Basu et al.,

2022) that have the potential to drive the next wave of computational technology

and artiﬁcial intelligence (Christensen et al., 2022; Frenkel et al., 2021; Shrestha

et al., 2022). Neuromorphic—that is, brain-like—computing systems imitate the

brain at the level of organizational principles (Indiveri and Liu, 2015), and often

also at the level of device physics by leveraging nonlinear phenomena in semi-

conductors (Chicca et al., 2014; Rubino et al., 2021) and other nanoscale devices

(Zidan et al., 2018; Markovi´c et al., 2020) for non-digital computation. The idea

of using nonlinear physical phenomena for non-digital computing has been ex-

plored for decades. Diﬀerent choices of underlying mathematical models lead

to diﬀerent deﬁnitions of what the concept of “computation” entails (Jaeger,

2021), and likely also inﬂuences the set of possible emergent innovations.

Here, we deﬁne neuromorphic computing (NC) systems as non–von Neu-

mann information-processing systems, the structure and function of which ei-

ther emulate or simulate the neuronal dynamics of brains—especially of somas,

but sometimes also synapses, dendrites, and axons—typically in the form of

spiking neural networks (SNNs) (Maass, 1997; Nunes et al., 2022; Wang et al.,

2022). NC systems open up new algorithmic spaces—through asynchronous

massive parallelism, sparse, event-driven activity, and co-location of memory

and processing (Indiveri and Liu, 2015)—and, in terms of energy-usage and la-

tency, oﬀer superior solutions to a range of brain-like computational problems

(Davies et al., 2021; Yin et al., 2021; St¨ockl and Maass, 2021; G¨oltz et al.,

2021; Rao et al., 2022). Furthermore, beyond cognitive applications, SNNs and

NC systems have also demonstrated potential for applications such as graph al-

gorithms, constrained optimization, random walks, partial-diﬀerential-equation

solving, signal processing, and algorithm composition (Aimone et al., 2022).

Consequently, there is a growing interest for NC technology within applica-

tion domains such as automotive technology, digitized industrial production

and monitoring, mobile devices, robotics, biosensing (such as brain–machine in-

terfaces and wearables), prosthetics, telecommunications-network (5G/6G) op-

timization, and space technology.

One challenge facing neuromorphic technology is that of integrating emerg-

ing diverse hardware systems, such as neuromorphic processors and quantum

computers, into a common computational environment (Vetter et al., 2018).

Such hardware systems are—due to performance constraints of existing compu-

tational hardware in, for instance, energy usage or processing speed—likely to

be increasingly included in computational ecosystems to facilitate or accelerate

particular types of computation (Shalf, 2020; Leiserson et al., 2020; Hamilton

et al., 2020). Fundamental trends in computer-architecture development indi-

cate that nearly all aspects of future high-performance computing architectures

will have substantially higher numbers of diverse and unconventional compo-

nents than past architectures (Becker et al., 2022), leading toward a period of

“extreme heterogeneity”. Consequently, neuromorphic processors are, in many

future use-cases, likely to be part of a broader, heterogeneous computational en-

vironment, rather than to be operated in isolation. Thus, there is a need for pro-

gramming models and abstractions, as well as interparadigmatic communication

principles and data models, that enable eﬀective integration of neuromorphic

hardware into large-scale systems of systems.

Here, we address the problem of large-scale adoption and integration of NC

systems into the present digital–computational infrastructure. We frame this

problem in terms of the following main challenges:

1. Communication: How to transcode information between neuromorphic

and digital systems?

2. Virtualization: How to interface neuromorphic devices and services in

distributed digitized systems?

3. Programming: How to eﬃciently program hybrid neuromorphic–digital

systems?

4. Testing and validation: How to reliably train and test the functionality

of such hybrid systems?

We outline the current landscape of NC technology from the perspective of

system integration—describing the most signiﬁcant qualities of NC systems as

compared to the fundamentally diﬀerent paradigm of conventional digital com-

puting (DC)1. Based on this description, we propose a microservice-based frame-

work for integration of NC systems. The framework consists of a neuromorphic-

system proxy, which provides virtualization and communication capabilities re-

quired in a distributed setting, in combination with a declarative programming

approach oﬀering engineering-process abstraction. We present established con-

cepts for programming, representation, and communication in distributed sys-

tems that could serve as a basis for the realization of this framework, and identify

directions for further research required to enable such integration.

2 Neuromorphic Systems

The ﬁeld of neuromorphic engineering dates back to the late 1980s (Mead, 1990,

2020), and originally dealt with the creation and use of sensing and processing

systems that imitate the brain at the level of structure and device physics. To-

day, the term “neuromorphic” has broadened, and “neuromorphic processors”

typically refer to hardware systems of diﬀerent architectures that are special-

ized for running SNNs. Neuromorphic hardware architectures thus range from

electronic emulation with analog circuitry or novel electronic devices to digital

1The terms “digital computing (DC) system” and “von Neumann computer” are used

interchangeably throughout this article.

systems specialized for massively parallel diﬀerential-equation solving for spik-

ing neuron models. However, as spiking neural networks (Maass, 1997; Nunes

et al., 2022; Wang et al., 2022) are inherently event-driven, asynchronous, time-

dependent, and highly parallel, all neuromorphic processors, by consequence,

diﬀer signiﬁcantly from von Neumann computers, as summarized in Table 1.

Table 1: Qualitative diﬀerences between von Neumann and neuromorphic computational architectures.

Architecture von Neumann Neuromorphic

Processing operations Sequential Massively parallel

Memory–processing organization Centralized, separated Distributed, colocated

Temporal organization Synchronous, clock-driven Asynchronous, event-driven

State qualities Discrete, static Continuous, dynamic

Programming method Sequential logic Structural SNN conﬁguration

Unit of communication Binary numbers Unary spike-events (spatiotemporal, sparse)

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

IntegrationofNeuromorphicAIinEvent-DrivenDistributedDigitizedSystems:ConceptsandResearchDirectionsMattiasNilsson*,OlovSchelen*,AndersLindgren*,UlfBodin*,CristinaPaniagua*,JerkerDelsing*,andFredrikSandin*October21,2022AbstractIncreasingcomplexityanddata-generationratesincyber-physicalsys-temsandthe...

展开>> 收起<<

Integration of Neuromorphic AI in Event-Driven Distributed Digitized Systems Concepts and Research Directions.pdf

共31页,预览5页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Integration of Neuromorphic AI in Event-Driven Distributed Digitized Systems Concepts and Research Directions

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: