Energy efficient 'in memory' computing to enable decentralised service workflow composition in support of multi-domain operations

Bent, Graham; Simpkin, Christopher; Taylor, Ian; Rahimi, Abbas; Karunaratne, Geethan; Sebastian, Abu; Millar, D. A.; Martens, Andreas; Roy, Kaushik

doi:10.1117/12.2586988

Cited by 4 publications

(17 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the SPIE21 11 paper it was demonstrated that HVs can be used to perform IoBT service workflow orchestration in typical MDO TacCIS environments. In these environments D is limited by the capability to efficiently exchange HVs over relatively low bandwidth communications networks.…”

Section: Bsc Vs Shv Bundling Capacitymentioning

confidence: 99%

“…To demonstrate how a VSA approach can be used to perform typical MDO tasks the SPIE21 11 paper described a communications re-planning task being performed in a typical MDO TacCIS environment where agile rapid communications re-planning is required. The demonstration showed how a radios and communications plans can be represented as VSA HVs and how by exchanging the HVs over a number of radio networks, the HVs can be used to discover radio's operating under a current communications plan and instruct them to switch to a required target communications plan.…”

Section: Using Shvs and Snn Circuits In An Mdo Scenariomentioning

confidence: 99%

“…In a previous paper presented at SPIE 2021 11 , a VSA was described that used an hierarchical BSC binding and bundling scheme [12][13][14] . This new approach to workflow orchestration was used to demonstrate how MDO sensor and service descriptions could be represented as HVs and how these HVs could in turn be bound and bundled into higher-level HVs that represent sensor-service workflows.…”

Section: Introductionmentioning

confidence: 99%

“…Although PCM devices can efficiently perform the vector matching they are not able to perform the other required VSA processing operations, namely binding, bundling and unbinding that are required to implement a full VSA solution. The motivation for this paper was to demonstrate the possibility to perform all of the VSA operations required to implement the hierarchical VSA scheme presented in the SPIE21 11 paper using an alternative technology based on emerging energy efficient SNN neuromorphic processing devices such as the IBM TrueNorth 16 and Intel Loihi 2 17 processors. Eliasmith 5 has demonstrated how spike-rate encoded SNN circuits can implement an HRR based VSA scheme for cognitive processing in order to perform the required convolution and deconvolution operations.…”

Section: Introductionmentioning

confidence: 99%

“…Section 3 describes how the Lahio model can be mapped to a time-to-spike SNN model and how all the basic VSA operations can be represented as SNN circuits. Section 4 summarises the MDO communications re-planning scenario presented in the SPIE21 11 paper. This scenario is used to illustrate how the basic SNN circuits can be combined to perform the more complex VSA functions required and compares the estimated energy requirements for SNN devices with those obtained for PCM devices performing similar functions.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Energy efficient spiking neural network neuromorphic processing to enable decentralised service workflow composition in support of multi-domain operations

Bent

Simpkin

Li³

et al. 2022

Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications IV

View full text Add to dashboard Cite

Section: Bsc Vs Shv Bundling Capacitymentioning

confidence: 99%

Section: Using Shvs and Snn Circuits In An Mdo Scenariomentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Energy efficient spiking neural network neuromorphic processing to enable decentralised service workflow composition in support of multi-domain operations

Bent

Simpkin

Li³

et al. 2022

Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications IV

View full text Add to dashboard Cite

Hyperdimensional Computing Using Time-To-Spike Neuromorphic Circuits

Bent

Simpkin

et al. 2022

2022 International Joint Conference on Neural Networks (IJCNN)

Self Cite

View full text Add to dashboard Cite

Vector Symbolic Architectures (VSA) can be used to encode complex objects, such as services and sensors, as hypervectors. Such hypervectors can be used to perform efficient distributed service discovery and workflow orchestration in communications constrained environments typical of the Internet of Things (IoT). In these environments, energy efficiency is of great importance. However, most hypervector representations use dense i.i.d element values and performing energy efficient hyperdimensional computing operations on such dense vectors is challenging. More recently, a sparse binary VSA scheme has been proposed based on a slot encoding having slots with bit positions per slot, in which only one bit per slot can be set. This paper shows for the first time that such sparse encoded hypervectors can be mapped into energy-efficient timeto-spike Spiking Neural Network (SNN) circuits, such that all the required VSA operations can be performed. Example VSA SNN circuits have been implemented in the Brian 2 SNN simulator, showing that all VSA binding, bundling, unbinding, and cleanup memory operations execute correctly. Based on these circuit implementations, estimates of the energy and processing time required to perform the different VSA operations on typical SNN neuromorphic devices are estimated. Recommendations for the design of future SNN neuromorphic processor hardware that can more efficiently perform VSA processing are also made.

show abstract

Semantic vector space mapping in edge of network multi-domain operations

Bent¹,

Summers-Stay

Preece

et al. 2023

Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications V

View full text Add to dashboard Cite

Artificial Intelligence applications are increasingly making use of vector embedding techniques to achieve impressive results in many application domains. Semantic Vector Spaces (SVS's) are constructed using semantic vector embedding techniques that learn vector representations of data across multiple domains. An important application area enabled by such techniques is the capability to represent software services and service workflows as semantic hypervectors. Previous work has shown how these hypervector representations have significant advantages over alternative schemes for decentralized service workflow construction, particularly in low communications bandwidth and energy constrained environments that are typical in multi-domain operations. SVS construction usually assumes that all the data required to construct the semantic vector space is available centrally. However, in multi-domain operations different partners may not be willing to share the training data necessary to construct a common multi-domain SVS. Hence semantic hypervectors representing similar services or workflows but constructed from different training data by different partners cannot be discovered and used. In this paper we focus on how it is possible to map semantic hypervectors between partner SVS's, so that complementary services and workflows developed and owned by different partners can be discovered and used to achieve mission goals. The paper describes techniques for generating the required mapping that require a minimum exchange of information between the different partners; demonstrate how it is possible to do this for semantic hypervectors that use different types of encoding (eg., real valued, binary, sparse slot-encoding); and illustrates how the mapping can be implemented in various multi-domain operational settings.

show abstract

Energy efficient 'in memory' computing to enable decentralised service workflow composition in support of multi-domain operations

Cited by 4 publications

References 18 publications

Energy efficient spiking neural network neuromorphic processing to enable decentralised service workflow composition in support of multi-domain operations

Energy efficient spiking neural network neuromorphic processing to enable decentralised service workflow composition in support of multi-domain operations

Hyperdimensional Computing Using Time-To-Spike Neuromorphic Circuits

Semantic vector space mapping in edge of network multi-domain operations

Contact Info

Product

Resources

About