Efficient Processing of Spatio-Temporal Data Streams With Spiking Neural Networks
Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to exploit these advantages. Although ANN-to-SNN conversion has...
Saved in:
Published in | Frontiers in neuroscience Vol. 14; p. 439 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
Switzerland
Frontiers Research Foundation
05.05.2020
Frontiers Media S.A |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to exploit these advantages. Although ANN-to-SNN conversion has achieved state-of-the-art accuracy for static image classification tasks, the following subtle but important difference in the way SNNs and ANNs integrate information over time makes the direct application of conversion techniques for sequence processing tasks challenging. Whereas all connections in SNNs have a certain propagation delay larger than zero, ANNs assign different roles to feed-forward connections, which immediately update all neurons within the same time step, and recurrent connections, which have to be rolled out in time and are typically assigned a delay of one time step. Here, we present a novel method to obtain highly accurate SNNs for sequence processing by modifying the ANN training before conversion, such that delays induced by ANN rollouts match the propagation delays in the targeted SNN implementation. Our method builds on the recently introduced framework of streaming rollouts, which aims for fully parallel model execution of ANNs and inherently allows for temporal integration by merging paths of different delays between input and output of the network. The resulting networks achieve state-of-the-art accuracy for multiple event-based benchmark datasets, including N-MNIST, CIFAR10-DVS, N-CARS, and DvsGesture, and through the use of spatio-temporal shortcut connections yield low-latency approximate network responses that improve over time as more of the input sequence is processed. In addition, our converted SNNs are consistently more energy-efficient than their corresponding ANNs. |
---|---|
AbstractList | Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to exploit these advantages. Although ANN-to-SNN conversion has achieved state-of-the-art accuracy for static image classification tasks, the following subtle but important difference in the way SNNs and ANNs integrate information over time makes the direct application of conversion techniques for sequence processing tasks challenging. Whereas all connections in SNNs have a certain propagation delay larger than zero, ANNs assign different roles to feed-forward connections, which immediately update all neurons within the same time step, and recurrent connections, which have to be rolled out in time and are typically assigned a delay of one time step. Here, we present a novel method to obtain highly accurate SNNs for sequence processing by modifying the ANN training before conversion, such that delays induced by ANN rollouts match the propagation delays in the targeted SNN implementation. Our method builds on the recently introduced framework of streaming rollouts, which aims for fully parallel model execution of ANNs and inherently allows for temporal integration by merging paths of different delays between input and output of the network. The resulting networks achieve state-of-the-art accuracy for multiple event-based benchmark datasets, including \nmnist{}, \cifdvs{}, \ncars{}, and \dvsgest{}, and through the use of spatio-temporal shortcut connections yield low-latency approximate network responses that improve over time as more of the input sequence is processed. In addition, our converted SNNs are consistently more energy-efficient than their corresponding ANNs. Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to exploit these advantages. Although ANN-to-SNN conversion has achieved state-of-the-art accuracy for static image classification tasks, the following subtle but important difference in the way SNNs and ANNs integrate information over time makes the direct application of conversion techniques for sequence processing tasks challenging. Whereas all connections in SNNs have a certain propagation delay larger than zero, ANNs assign different roles to feed-forward connections, which immediately update all neurons within the same time step, and recurrent connections, which have to be rolled out in time and are typically assigned a delay of one time step. Here, we present a novel method to obtain highly accurate SNNs for sequence processing by modifying the ANN training before conversion, such that delays induced by ANN rollouts match the propagation delays in the targeted SNN implementation. Our method builds on the recently introduced framework of streaming rollouts, which aims for fully parallel model execution of ANNs and inherently allows for temporal integration by merging paths of different delays between input and output of the network. The resulting networks achieve state-of-the-art accuracy for multiple event-based benchmark datasets, including N-MNIST, CIFAR10-DVS, N-CARS, and DvsGesture, and through the use of spatio-temporal shortcut connections yield low-latency approximate network responses that improve over time as more of the input sequence is processed. In addition, our converted SNNs are consistently more energy-efficient than their corresponding ANNs. Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to exploit these advantages. Although ANN-to-SNN conversion has achieved state-of-the-art accuracy for static image classification tasks, the following subtle but important difference in the way SNNs and ANNs integrate information over time makes the direct application of conversion techniques for sequence processing tasks challenging. Whereas all connections in SNNs have a certain propagation delay larger than zero, ANNs assign different roles to feed-forward connections, which immediately update all neurons within the same time step, and recurrent connections, which have to be rolled out in time and are typically assigned a delay of one time step. Here, we present a novel method to obtain highly accurate SNNs for sequence processing by modifying the ANN training before conversion, such that delays induced by ANN rollouts match the propagation delays in the targeted SNN implementation. Our method builds on the recently introduced framework of streaming rollouts, which aims for fully parallel model execution of ANNs and inherently allows for temporal integration by merging paths of different delays between input and output of the network. The resulting networks achieve state-of-the-art accuracy for multiple event-based benchmark datasets, including N-MNIST , CIFAR10-DVS , N-CARS , and DvsGesture , and through the use of spatio-temporal shortcut connections yield low-latency approximate network responses that improve over time as more of the input sequence is processed. In addition, our converted SNNs are consistently more energy-efficient than their corresponding ANNs. |
Author | Pfeil, Thomas Chicca, Elisabetta Kugele, Alexander Pfeiffer, Michael |
AuthorAffiliation | 1 Faculty of Technology and Center of Cognitive Interaction Technology (CITEC), Bielefeld University , Bielefeld , Germany 2 Bosch Center for Artificial Intelligence , Renningen , Germany |
AuthorAffiliation_xml | – name: 1 Faculty of Technology and Center of Cognitive Interaction Technology (CITEC), Bielefeld University , Bielefeld , Germany – name: 2 Bosch Center for Artificial Intelligence , Renningen , Germany |
Author_xml | – sequence: 1 givenname: Alexander surname: Kugele fullname: Kugele, Alexander organization: Bosch Center for Artificial Intelligence, Renningen, Germany – sequence: 2 givenname: Thomas surname: Pfeil fullname: Pfeil, Thomas organization: Bosch Center for Artificial Intelligence, Renningen, Germany – sequence: 3 givenname: Michael surname: Pfeiffer fullname: Pfeiffer, Michael organization: Bosch Center for Artificial Intelligence, Renningen, Germany – sequence: 4 givenname: Elisabetta surname: Chicca fullname: Chicca, Elisabetta organization: Faculty of Technology and Center of Cognitive Interaction Technology (CITEC), Bielefeld University, Bielefeld, Germany |
BackLink | https://www.ncbi.nlm.nih.gov/pubmed/32431592$$D View this record in MEDLINE/PubMed |
BookMark | eNpdkc9vFCEUx4lpY3_o3ZOZxEsvs2WAGeBiYmqrTZqqaY3eCMM8tmxnYIWZNv73Mrt1Yz1B4PM-8N73CO354AGhNxVeUCrkqfXOpwXBBC8wZlS-QIdV05CS1fTn3m7PxAE6SmmFcUMEIy_RASWMVrUkh-jbubXOOPBj8TUGAyk5vyyCLW7WenShvIVhHaLui4961MXNGEEPqfjhxrtMuPsZvoZpBq5hfAzxPr1C-1b3CV4_rcfo-8X57dnn8urLp8uzD1elYRKPZUUb2Una8pYLXDNL6g5bY7oOWC0ItJxWuOWGd1a2nHCOa2mklLkl2jTWtPQYXW69XdArtY5u0PG3CtqpzUGIS6Xj6EwPqmlBM46FtkSyRgutm5Yz0PkRAR2rsuv91rWe2gE6k8eRW3omfX7j3Z1ahgfFScUEnwUnT4IYfk2QRjW4ZKDvtYcwJUUYrikmkpOMvvsPXYUp-jwqRagUVGBWyUzhLWViSCmC3X2mwmrOXm2yV3P2apN9Lnn7bxO7gr9h0z_Eoa0- |
CitedBy_id | crossref_primary_10_3389_fnins_2023_1261543 crossref_primary_10_1162_neco_a_01499 crossref_primary_10_3389_fnins_2023_1213720 crossref_primary_10_1109_TNNLS_2021_3073016 crossref_primary_10_3389_fnins_2022_1079357 crossref_primary_10_3389_fnins_2022_759900 crossref_primary_10_1016_j_neunet_2023_10_056 crossref_primary_10_1016_j_neucom_2023_126485 crossref_primary_10_1016_j_neunet_2022_02_006 crossref_primary_10_3389_fnins_2022_923587 crossref_primary_10_1088_2634_4386_ac8bef crossref_primary_10_1016_j_ifacol_2022_09_156 crossref_primary_10_1016_j_neucom_2021_08_035 crossref_primary_10_1088_2634_4386_ac970d crossref_primary_10_1145_3571155 crossref_primary_10_1109_TCDS_2023_3308347 crossref_primary_10_3390_make5030052 crossref_primary_10_1021_acsami_0c15519 crossref_primary_10_1002_adts_202200226 crossref_primary_10_1016_j_envsoft_2023_105851 crossref_primary_10_1016_j_ins_2023_119170 crossref_primary_10_1088_1741_2552_aceca3 crossref_primary_10_1109_ACCESS_2022_3209671 crossref_primary_10_1002_aisy_202400265 crossref_primary_10_1109_ACCESS_2023_3278098 crossref_primary_10_1016_j_neunet_2024_106121 crossref_primary_10_1016_j_knosys_2023_111024 crossref_primary_10_1016_j_ins_2024_120136 crossref_primary_10_1007_s13748_024_00313_4 crossref_primary_10_1126_sciadv_adl2767 crossref_primary_10_1002_cta_3075 |
Cites_doi | 10.3389/fnins.2017.00309 10.1126/science.1254642 10.1038/srep40703 10.3389/fnins.2015.00437 10.1109/CVPR.2018.00291 10.1016/j.neunet.2019.09.005 10.1016/j.mejo.2015.10.012 10.1109/CVPR.2018.00186 10.1109/ICIP.2017.8296389 10.1109/ICCV.2019.00573 10.1109/WACV.2019.00199 10.3389/fnins.2016.00508 10.1109/TPAMI.2013.71 10.1109/ISCAS.2010.5537150 10.1109/CVPR.2015.7298664 10.1016/j.conb.2010.03.007 10.3389/fnins.2018.00331 10.3389/fnins.2016.00594 10.3389/fnins.2017.00682 10.1007/s11263-014-0788-3 10.1109/5.58337 10.1109/CVPR.2017.243 10.3389/fnins.2015.00141 10.1609/aaai.v33i01.33011311 10.1109/TMI.2018.2823338 10.3389/fnins.2012.00032 10.1017/CBO9781107447615 10.1109/ICRC.2016.7738691 10.1109/MM.2018.112130359 10.1109/ISCAS.2010.5536970 10.1109/MC.2019.2903009 10.1109/TC.2012.142 10.1109/CVPR.2019.00398 10.1162/NECO_a_00882 10.3389/fnins.2013.00178 10.3389/fnins.2018.00774 10.3389/fnins.2018.00891 |
ContentType | Journal Article |
Copyright | Copyright © 2020 Kugele, Pfeil, Pfeiffer and Chicca. 2020. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. Copyright © 2020 Kugele, Pfeil, Pfeiffer and Chicca. 2020 Kugele, Pfeil, Pfeiffer and Chicca |
Copyright_xml | – notice: Copyright © 2020 Kugele, Pfeil, Pfeiffer and Chicca. – notice: 2020. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. – notice: Copyright © 2020 Kugele, Pfeil, Pfeiffer and Chicca. 2020 Kugele, Pfeil, Pfeiffer and Chicca |
DBID | NPM AAYXX CITATION 3V. 7XB 88I 8FE 8FH 8FK ABUWG AFKRA AZQEC BBNVY BENPR BHPHI CCPQU DWQXO GNUQQ HCIFZ LK8 M2P M7P PIMPY PQEST PQQKQ PQUKI PRINS Q9U 7X8 5PM DOA |
DOI | 10.3389/fnins.2020.00439 |
DatabaseName | PubMed CrossRef ProQuest Central (Corporate) ProQuest Central (purchase pre-March 2016) Science Database (Alumni Edition) ProQuest SciTech Collection ProQuest Natural Science Collection ProQuest Central (Alumni) (purchase pre-March 2016) ProQuest Central (Alumni) ProQuest Central UK/Ireland ProQuest Central Essentials Biological Science Collection AUTh Library subscriptions: ProQuest Central ProQuest Natural Science Collection ProQuest One Community College ProQuest Central ProQuest Central Student SciTech Premium Collection (Proquest) (PQ_SDU_P3) Biological Sciences ProQuest Science Journals Biological Science Database Publicly Available Content Database ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China ProQuest Central Basic MEDLINE - Academic PubMed Central (Full Participant titles) DOAJ Directory of Open Access Journals |
DatabaseTitle | PubMed CrossRef Publicly Available Content Database ProQuest Science Journals (Alumni Edition) ProQuest Central Student ProQuest Biological Science Collection ProQuest Central Basic ProQuest Central Essentials ProQuest Science Journals ProQuest One Academic Eastern Edition ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest Natural Science Collection Biological Science Database ProQuest SciTech Collection ProQuest Central China ProQuest Central ProQuest One Academic UKI Edition Natural Science Collection ProQuest Central Korea Biological Science Collection ProQuest One Academic ProQuest Central (Alumni) MEDLINE - Academic |
DatabaseTitleList | Publicly Available Content Database PubMed MEDLINE - Academic |
Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: BENPR name: AUTh Library subscriptions: ProQuest Central url: https://www.proquest.com/central sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Anatomy & Physiology |
EISSN | 1662-453X |
EndPage | 439 |
ExternalDocumentID | oai_doaj_org_article_6bea4708af2946a8aa6b74eab738ed41 10_3389_fnins_2020_00439 32431592 |
Genre | Journal Article |
GrantInformation_xml | – fundername: Deutsche Forschungsgemeinschaft – fundername: Horizon 2020 |
GroupedDBID | --- 29H 2WC 53G 5GY 5VS 88I 8FE 8FH 9T4 AAFWJ ABUWG ACGFO ACGFS ACXDI ADRAZ AEGXH AENEX AFKRA AFPKN AIAGR ALMA_UNASSIGNED_HOLDINGS AZQEC BBNVY BENPR BHPHI BPHCQ C1A CCPQU CS3 DIK DU5 DWQXO E3Z EBS EJD EMOBN F5P FRP GNUQQ GROUPED_DOAJ GX1 HCIFZ HYE IAO IEA IHR ISR ITC KQ8 LK8 M2P M48 M7P M~E NPM O5R O5S OK1 P2P PIMPY PQQKQ PROAC RNS RPM W2D AAYXX CITATION PGMZT 3V. 7XB 8FK PQEST PQUKI PRINS Q9U 7X8 5PM |
ID | FETCH-LOGICAL-c490t-1369d93b7b78054f25d0fccdde4582eb7310b7c7df9b7277059c999166366fcb3 |
IEDL.DBID | RPM |
ISSN | 1662-4548 1662-453X |
IngestDate | Thu Jul 04 21:11:18 EDT 2024 Tue Sep 17 21:27:05 EDT 2024 Sat Aug 17 05:45:23 EDT 2024 Wed Sep 25 00:51:54 EDT 2024 Fri Aug 23 03:26:00 EDT 2024 Fri Mar 29 01:43:54 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Keywords | spiking neural networks efficient inference event-based vision neuromorphic computing sequence processing |
Language | English |
License | Copyright © 2020 Kugele, Pfeil, Pfeiffer and Chicca. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c490t-1369d93b7b78054f25d0fccdde4582eb7310b7c7df9b7277059c999166366fcb3 |
Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 This article was submitted to Neuromorphic Engineering, a section of the journal Frontiers in Neuroscience Reviewed by: Jim Harkin, Ulster University, United Kingdom; Guoqi Li, Tsinghua University, China Edited by: Kaushik Roy, Purdue University, United States |
OpenAccessLink | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7214871/ |
PMID | 32431592 |
PQID | 2398380419 |
PQPubID | 4424402 |
PageCount | 1 |
ParticipantIDs | doaj_primary_oai_doaj_org_article_6bea4708af2946a8aa6b74eab738ed41 pubmedcentral_primary_oai_pubmedcentral_nih_gov_7214871 proquest_miscellaneous_2405302972 proquest_journals_2398380419 crossref_primary_10_3389_fnins_2020_00439 pubmed_primary_32431592 |
PublicationCentury | 2000 |
PublicationDate | 2020-05-05 |
PublicationDateYYYYMMDD | 2020-05-05 |
PublicationDate_xml | – month: 05 year: 2020 text: 2020-05-05 day: 05 |
PublicationDecade | 2020 |
PublicationPlace | Switzerland |
PublicationPlace_xml | – name: Switzerland – name: Lausanne |
PublicationTitle | Frontiers in neuroscience |
PublicationTitleAlternate | Front Neurosci |
PublicationYear | 2020 |
Publisher | Frontiers Research Foundation Frontiers Media S.A |
Publisher_xml | – name: Frontiers Research Foundation – name: Frontiers Media S.A |
References | DeBole (B6) 2019; 52 Bai (B2) 2018 Li (B19) 2017; 11 Furber (B12) 2013; 62 Neftci (B24) 2019 Orchard (B26) 2015; 9 Gallego (B13) 2019 Tompson (B39) 2015 Zhang (B46) 2018; 37 Wu (B44) 2018; 12 O'Connor (B25) 2013; 7 Esser (B9) 2015 van den Oord (B40) Fischer (B11) 2018 Huang (B17) 2017 Schemmel (B35) 2010 Thakur (B38) 2018; 12 Werbos (B43) 1990; 78 Gehrig (B14) 2019 Liu (B20) 2010; 20 Wu (B45) 2019 Osswald (B27) 2017; 7 Rieke (B33) 1999 Pfeiffer (B29) 2018; 12 Huang (B16) 2018 Sironi (B37) 2018 Merolla (B23) 2014; 345 Shrestha (B36) 2018 Pérez-Carrasco (B28) 2013; 35 Qiao (B31) 2015; 9 Rueckauer (B34) 2017; 11 Wang (B42) 2019 Cao (B3) 2015; 113 Zhu (B47) 2017 Martí (B22) 2015; 28 Davies (B5) 2018; 38 Diehl (B8) 2016 Posch (B30) 2010 Rebecq (B32) 2019 Farabet (B10) 2012; 6 Manolopoulos (B21) 2016; 49 Deng (B7) 2020; 121 Amir (B1) 2017 Gerstner (B15) 2014 van den Oord (B41) Lee (B18) 2016; 10 Clady (B4) 2017; 10 |
References_xml | – volume: 11 start-page: 309 year: 2017 ident: B19 article-title: Cifar10-dvs: an event-stream dataset for object classification publication-title: Front. Neurosci doi: 10.3389/fnins.2017.00309 contributor: fullname: Li – volume: 345 start-page: 668 year: 2014 ident: B23 article-title: A million spiking-neuron integrated circuit with a scalable communication network and interface publication-title: Science doi: 10.1126/science.1254642 contributor: fullname: Merolla – volume: 7 start-page: 40703 year: 2017 ident: B27 article-title: A spiking neural network model of 3d perception for event-based neuromorphic stereo vision systems publication-title: Sci. Rep doi: 10.1038/srep40703 contributor: fullname: Osswald – volume: 9 start-page: 437 year: 2015 ident: B26 article-title: Converting static image datasets to spiking neuromorphic datasets using saccades publication-title: Front. Neurosci doi: 10.3389/fnins.2015.00437 contributor: fullname: Orchard – volume-title: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) year: 2018 ident: B16 article-title: Condensenet: an efficient densenet using learned group convolutions doi: 10.1109/CVPR.2018.00291 contributor: fullname: Huang – volume: 121 start-page: 294 year: 2020 ident: B7 article-title: Rethinking the performance comparison between SNNs and ANNs publication-title: Neural Netw doi: 10.1016/j.neunet.2019.09.005 contributor: fullname: Deng – volume: 49 start-page: 10 year: 2016 ident: B21 article-title: An efficient multiple precision floating-point multiply-add fused unit publication-title: Microelectron. J doi: 10.1016/j.mejo.2015.10.012 contributor: fullname: Manolopoulos – volume-title: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) year: 2018 ident: B37 article-title: HATS: Histograms of averaged time surfaces for robust event-based object classification doi: 10.1109/CVPR.2018.00186 contributor: fullname: Sironi – start-page: 790 volume-title: 2017 IEEE International Conference on Image Processing (ICIP) year: 2017 ident: B47 article-title: Densenet for dense flow doi: 10.1109/ICIP.2017.8296389 contributor: fullname: Zhu – volume-title: The IEEE International Conference on Computer Vision (ICCV) year: 2019 ident: B14 article-title: End-to-end learning of representations for asynchronous event-based data doi: 10.1109/ICCV.2019.00573 contributor: fullname: Gehrig – volume-title: CoRR ident: B40 article-title: Wavenet: A generative model for raw audio contributor: fullname: van den Oord – start-page: 1826 volume-title: 2019 IEEE Winter Conference on Applications of Computer Vision (WACV) year: 2019 ident: B42 article-title: Space-time event clouds for gesture recognition: From RGB cameras to event cameras doi: 10.1109/WACV.2019.00199 contributor: fullname: Wang – volume: 10 start-page: 508 year: 2016 ident: B18 article-title: Training deep spiking neural networks using backpropagation publication-title: Front. Neurosci doi: 10.3389/fnins.2016.00508 contributor: fullname: Lee – volume: 35 start-page: 2706 year: 2013 ident: B28 article-title: Mapping from frame-driven to frame-free event-driven vision systems by low-rate rate coding and coincidence processing-application to feedforward convnets publication-title: IEEE Trans. Pattern Anal. Mach. Intell doi: 10.1109/TPAMI.2013.71 contributor: fullname: Pérez-Carrasco – start-page: 4039 volume-title: Advances in Neural Information Processing Systems 31 year: 2018 ident: B11 article-title: The streaming rollout of deep networks-towards fully model-parallel execution contributor: fullname: Fischer – start-page: 2430 year: 2010 ident: B30 article-title: High-DR frame-free PWM imaging with asynchronous AER intensity encoding and focal-plane temporal redundancy suppression publication-title: Proceedings of 2010 IEEE International Symposium on Circuits and Systems doi: 10.1109/ISCAS.2010.5537150 contributor: fullname: Posch – volume-title: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) year: 2015 ident: B39 article-title: Efficient object localization using convolutional networks doi: 10.1109/CVPR.2015.7298664 contributor: fullname: Tompson – start-page: 1412 volume-title: Advances in Neural Information Processing Systems 31 year: 2018 ident: B36 article-title: Slayer: spike layer error reassignment in time contributor: fullname: Shrestha – start-page: 1117 volume-title: Advances in Neural Information Processing Systems year: 2015 ident: B9 article-title: Backpropagation for energy-efficient neuromorphic computing contributor: fullname: Esser – volume: 20 start-page: 288 year: 2010 ident: B20 article-title: Neuromorphic sensory systems publication-title: Curr. Opin. Neurobiol doi: 10.1016/j.conb.2010.03.007 contributor: fullname: Liu – volume: 12 start-page: 331 year: 2018 ident: B44 article-title: Spatio-temporal backpropagation for training high-performance spiking neural networks publication-title: Front. Neurosci doi: 10.3389/fnins.2018.00331 contributor: fullname: Wu – volume: 10 start-page: 594 year: 2017 ident: B4 article-title: A motion-based feature for event-based pattern recognition publication-title: Front. Neurosci doi: 10.3389/fnins.2016.00594 contributor: fullname: Clady – volume: 11 start-page: 682 year: 2017 ident: B34 article-title: Conversion of continuous-valued deep networks to efficient event-driven networks for image classification publication-title: Front. Neurosci doi: 10.3389/fnins.2017.00682 contributor: fullname: Rueckauer – volume-title: Spikes: Exploring the Neural Code year: 1999 ident: B33 contributor: fullname: Rieke – volume: 113 start-page: 54 year: 2015 ident: B3 article-title: Spiking deep convolutional neural networks for energy-efficient object recognition publication-title: Int. J. Comput. Vis doi: 10.1007/s11263-014-0788-3 contributor: fullname: Cao – volume: 78 start-page: 1550 year: 1990 ident: B43 article-title: Backpropagation through time: what it does and how to do it publication-title: Proc. IEEE doi: 10.1109/5.58337 contributor: fullname: Werbos – volume-title: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) year: 2017 ident: B17 article-title: Densely connected convolutional networks doi: 10.1109/CVPR.2017.243 contributor: fullname: Huang – start-page: 7243 volume-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition year: 2017 ident: B1 article-title: A low power, fully event-based gesture recognition system contributor: fullname: Amir – volume-title: Surrogate Gradient Learning in Spiking Neural Networks year: 2019 ident: B24 contributor: fullname: Neftci – volume: 9 start-page: 141 year: 2015 ident: B31 article-title: A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128k synapses publication-title: Front. Neurosci doi: 10.3389/fnins.2015.00141 contributor: fullname: Qiao – volume-title: CoRR ident: B41 article-title: Pixel recurrent neural networks contributor: fullname: van den Oord – volume-title: Proceedings of the AAAI Conference on Artificial Intelligence year: 2019 ident: B45 article-title: Direct training of spiking neural networks: faster, larger, better doi: 10.1609/aaai.v33i01.33011311 contributor: fullname: Wu – volume: 37 start-page: 1407 year: 2018 ident: B46 article-title: A sparse-view ct reconstruction method based on combination of densenet and deconvolution publication-title: IEEE Trans. Med. Imaging doi: 10.1109/TMI.2018.2823338 contributor: fullname: Zhang – volume: 6 start-page: 32 year: 2012 ident: B10 article-title: Comparison between frame-constrained fix-pixel-value and frame-free spiking-dynamic-pixel convnets for visual processing publication-title: Front. Neurosci doi: 10.3389/fnins.2012.00032 contributor: fullname: Farabet – volume-title: Neuronal Dynamics: From Single Neurons to Networks and Models of Cognition year: 2014 ident: B15 doi: 10.1017/CBO9781107447615 contributor: fullname: Gerstner – start-page: 1 volume-title: 2016 IEEE International Conference on Rebooting Computing (ICRC) year: 2016 ident: B8 article-title: Conversion of artificial recurrent neural networks to spiking neural networks for low-power neuromorphic hardware doi: 10.1109/ICRC.2016.7738691 contributor: fullname: Diehl – volume: 38 start-page: 82 year: 2018 ident: B5 article-title: Loihi: a neuromorphic manycore processor with on-chip learning publication-title: IEEE Micro doi: 10.1109/MM.2018.112130359 contributor: fullname: Davies – start-page: 1947 volume-title: Proceedings of 2010 IEEE International Symposium on Circuits and Systems year: 2010 ident: B35 article-title: A wafer-scale neuromorphic hardware system for large-scale neural modeling doi: 10.1109/ISCAS.2010.5536970 contributor: fullname: Schemmel – volume: 52 start-page: 20 year: 2019 ident: B6 article-title: Truenorth: accelerating from zero to 64 million neurons in 10 years publication-title: Computer doi: 10.1109/MC.2019.2903009 contributor: fullname: DeBole – volume: 62 start-page: 2454 year: 2013 ident: B12 article-title: Overview of the spinnaker system architecture publication-title: IEEE Trans. Comput doi: 10.1109/TC.2012.142 contributor: fullname: Furber – year: 2019 ident: B32 article-title: Events-to-video: bringing modern computer vision to event cameras publication-title: CoRR doi: 10.1109/CVPR.2019.00398 contributor: fullname: Rebecq – start-page: abs/1803.01271 year: 2018 ident: B2 article-title: An empirical evaluation of generic convolutional and recurrent networks for sequence modeling publication-title: CoRR contributor: fullname: Bai – volume-title: CoRR year: 2019 ident: B13 article-title: Event-based vision: a survey contributor: fullname: Gallego – volume: 28 start-page: 2011 year: 2015 ident: B22 article-title: Energy-efficient neuromorphic classifiers publication-title: Neural Comput doi: 10.1162/NECO_a_00882 contributor: fullname: Martí – volume: 7 start-page: 178 year: 2013 ident: B25 article-title: Real-time classification and sensor fusion with a spiking deep belief network publication-title: Front. Neurosci doi: 10.3389/fnins.2013.00178 contributor: fullname: O'Connor – volume: 12 start-page: 774 year: 2018 ident: B29 article-title: Deep learning with spiking neurons: opportunities and challenges publication-title: Front. Neurosci doi: 10.3389/fnins.2018.00774 contributor: fullname: Pfeiffer – volume: 12 start-page: 891 year: 2018 ident: B38 article-title: Large-scale neuromorphic spiking array processors: a quest to mimic the brain publication-title: Front. Neurosci doi: 10.3389/fnins.2018.00891 contributor: fullname: Thakur |
SSID | ssj0062842 |
Score | 2.5786128 |
Snippet | Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods... |
SourceID | doaj pubmedcentral proquest crossref pubmed |
SourceType | Open Website Open Access Repository Aggregation Database Index Database |
StartPage | 439 |
SubjectTerms | Classification efficient inference event-based vision Firing pattern Information processing Latency Neural networks neuromorphic computing Neurons Neuroscience sequence processing spiking neural networks |
SummonAdditionalLinks | – databaseName: DOAJ Directory of Open Access Journals dbid: DOA link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Ba9swFBYlp11Kt7Rr1myoUAo9mNqWLEXHbGspgxVKG5abeZIlGlic0iaH_fu9JzkhGYVderWEJX_Pkr4nPX2PsTPIq8bZ0mZ5BU0mA5QZWNrIN5VHdm9VKem-889bdTORP6bVdCvVF8WEJXngBNylsh6kzkcQSiMVjACU1dKD1WLkG5kcn6JaO1NpDlY46ZbpUBJdMHMZ2llL2twlxXFJSgy-tQhFrf7XCOa_cZJbC8_1AdvvGCMfp56-Z3u-_cD64xa95fkffs5jDGfcHO-zu6uoCIEv4t0NAFyZ-CLw-xg4nT0kIarf_DssgdOJNMxf-K_Z8hFrzGjXnJNaB1a4TeHhL4dscn318O0m65ImZE6anFLLK9MYYbWlbAUylFWTB-dwFqMTMo-wFbnVTjfBWOQuGumViyRRCaWCs-KI9dpF648Z94WzoIM1oXKyAQuuCUUoPC7zYEQoB-xijWL9lLQxavQpCPE6Il4T4nVEfMC-EsybeqRqHR-grevO1vX_bD1gw7WR6m6oYRvCjASpKGEbp5tiHCR08gGtX6ywDtJSQWm6sM8fk003PUFGKZDTYYnesfZOV3dL2tljFOJG7xn9veLTW3zbCXtHaMVYymrIesvnlf-MfGdpv8Rf-y_YZQH8 priority: 102 providerName: Directory of Open Access Journals – databaseName: AUTh Library subscriptions: ProQuest Central dbid: BENPR link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1LT9wwELYoXLhUFFpYSitXQpV6iDYPx16fKmgXrSqx6gMEt2j8KiuVhEI49N93xklWLEJc11Z2NLbH3zz8DWOHkJbOmtwkaQkuEQHyBAwF8nXpEd0bmQt673w6l7Nz8e2yvFxjs-EtDJVVDjYxGmrXWIqRj4mnriCyHD0GQ1EA244_3_xNqH8U5Vn7Zhov2EaeCUrYbhxP599_DlZZohmOmU9Jr4QQpncpS3TQ9DjUi5qYu3Oq8hLUNvzBFRWZ_J-Cn4-rKB9cSydb7GWPJ_lRtwFesTVfb7Odoxp96et__COPFZ4xdL7DfkwjXwR-iPfvA_De4k3gv2JZdXLW0VT94V-hBU75ari-4xeL9gpnLCimzonLAyfMu-Lxu9fs_GR69mWW9C0VEit0So3npXa6MMpQLwMR8tKlwVq0cZQ_80Yh2jPKKhe0QWSjEHzZCCFlIWWwpnjD1uum9nuM-8waUMHoUFrhwIB1IQuZRxAAugj5iH0atFjddMwZFXocpPEqarwijVdR4yN2TGpeziPO6_hDc_u76o9QJY0HodIJhFwLCRMAaZTwgEJPvBPZiB0Mi1T1BxH_Y7ltRuzDchiPEOVFoPbNPc5B0FpQEy-Uebdb06UkiDcLRHw4olZWe0XU1ZF6cRVputG3Rm8w239erLdsk_QQayjLA7be3t77d4hzWvO-38L_AQKS_5s priority: 102 providerName: ProQuest – databaseName: Scholars Portal Journals (Open Access) dbid: M48 link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3faxQxEB60gvSlqPXHtVUiiODD6m42m9w-iFRtKUILYg_7FibZxB60e-3dFdr_3pns3unJPfm6yWbDZLLzTWbyDcAbzKvGO-myvMImUxFlho4P8usqELp3Wiq-73x8oo9G6ttZdfbnenQvwNla147rSY2mF-9vr-8-0Yb_yB4n2dsPsR23zLwtOUuLDOx9eCBVqVjfj9UypqDpR5xin5rvCRFQ74KWa0fYhIcENEoy9XLFXiVa_3VY9N-Uyr9s1OEj2OrBpdjvtOEx3AvtE9jeb8mxvrwTb0VK90zn6Nvw_SCRR9BAor8sQEZMTKL4kXKss9OOs-pCfMU5Cg5e4-VM_BzPz6nHmA_YBRN7UIeTLpN89hRGhwenX46yvr5C5lWdcxV6XTd16YzjwgYqyqrJo_f0w-NgWnCGoJ8z3jSxdgRzDCExn_CkLrWO3pXPYKOdtOEFiFB4hya6OlZeNejQN7GIRSBEgHUZ5QDeLaRorzoaDUvuBwvfJuFbFr5Nwh_AZxbzsh8TYKcHk-kv2-8nq11AZfIhRlkrjUNE7YwKSJMehkYVA9hbLJJdKJVlrsOSCZfoG6-XzbSfOEiCbZjcUB9CsCVX9KI5P-_WdDmThU4MwKys9spUV1va8Xni7CZHm1zDYue_39yFTRZRyrWs9mBjPr0JLwkPzd2rpOa_AazxC3w priority: 102 providerName: Scholars Portal |
Title | Efficient Processing of Spatio-Temporal Data Streams With Spiking Neural Networks |
URI | https://www.ncbi.nlm.nih.gov/pubmed/32431592 https://www.proquest.com/docview/2398380419/abstract/ https://search.proquest.com/docview/2405302972 https://pubmed.ncbi.nlm.nih.gov/PMC7214871 https://doaj.org/article/6bea4708af2946a8aa6b74eab738ed41 |
Volume | 14 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Nb9QwEB21RUK9IKB8LJSVkRASh3ST2LHjY1u2qpB2VaAVvUW2Y9NI3WzVbg_9951xkhWLOHHJYeNNRmPH88bz_AzwyaRF7Wxuk7QwdSKCyRNjaSFfFx7RvZW5oP3Os7k8vRDfLovLLSiGvTCRtO9sc9BeLw7a5ipyK28WbjLwxCZns2PMWhBnZ5Nt2FacDyl6N_1KnG9jiVPSdiDE411tEjMxPQlt05JEd050LozEu_AU8QTHiJ5vhKWo3v8vyPk3c_KPUHTyHJ71GJIddra-gC3fvoS9wxbz58UD-8wiqzMul-_B92nUiMAHsX5PAMYqtgzsZ6RSJ-edNNU1-2pWhlGN2izu2K9mdYUtGlpHZ6TfgQ3mHWH87hVcnEzPj0-T_hiFxAmd0mHzUteaW2Xp_AIR8qJOg3M4r1HNzFuFCM8qp-qgLaIZhYDLRdgouZTBWf4adtpl698C85mzRgWrQ-FEbaxxdchC5jHwG81DPoIvgxerm04to8Isg5xfRedX5PwqOn8ER-TmdTvSuY4_LG9_V31vV9J6I1RampBrIU1pjLRKeINGl74W2Qj2h06q-o8P38F1yUlXCd_xcX0bPxuqhZjWL--xDQJVTgd3oc1vuj5dWzKMiRGojd7eMHXzDo7UKM3dj8x3__3P97BLLoqUymIfdla39_4Dwp6VHcOTo-n87Mc4LhvgdSbKcRz6jwBhBts |
link.rule.ids | 230,315,733,786,790,870,891,2115,21416,24346,27957,27958,33779,33780,43840,53827,53829,74659 |
linkProvider | National Library of Medicine |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1LbxMxEB5BeoALAsojpYCREBKHVffhteMTaiFVgDbikYreLD9pJLrbNumBf8-MdxM1CHFdW7uj8Xr8zcPfALw2ee2dLW2W18ZnPJoyM5YC-aoOiO6tKDnddz6eiskJ_3Ran_YBt0VfVrmyiclQ-9ZRjHyPeOoqIstR7y4uM-oaRdnVvoXGbdjiFboqA9g6GE-_fFvZYoHGN-U7Bd0NQnDeJSrRLVN7sZk3xNddUm0Xp2bhNw6mxN__L9D5d-3kjcPo8D7c61Ek2--W_QHcCs1D2N5v0IM-_83esFTXmQLm2_B1nFgi8EWsvxWApxVrI_ueiqmzWUdO9Yt9MEvDKEttzhfsx3x5hjPmFElnxOCBE6ZdyfjiEZwcjmfvJ1nfSCFzXOXUbl4oryorLXUw4LGsfR6dQ8tGWbNgJWI8K530UVnEMxIhl0vAUVRCRGerxzBo2iY8BRYKZ42MVsXacW-scT4WsQh49BtVxXIIb1da1BcdX4ZGP4M0rpPGNWlcJ40P4YDUvJ5HTNfpQXv1U_cbRwsbDJf5yMRScWFGxggreTAo9Ch4Xgxhd7VIut9--I31zzKEV-th3DiUDTFNaK9xDkLVilp3ocxPujVdS4Ios0KchyNyY7U3RN0caeZniZwbPWr0AYud_4v1Eu5MZsdH-ujj9PMzuEs6SVWU9S4MllfX4TkinaV90f_OfwAAW_7O |
linkToPdf | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1LbxMxELaglRAX1FIeaQsYCSFxWHUfXjs-oT4SlVdUoBW9WWOvTSPR3bZJD_x7ZrxO1CDENbY2o_F4_I1n_A1jbyCvG2dLm-U1NJkIUGZg6SJf1x7RvZWloPfOXyby-Ex8PK_PU_3TLJVVLnxidNRN5-iOfI946ioiy9F7IZVFnByN319dZ9RBijKtqZ3GfbauhKzRwtcPRpOTbwu_LNERx9ynpHdCCNT7pCWGaPjFdtoSd3dJdV6CGoffOaQil_-_AOjfdZR3DqbxBnuUECXf701gk93z7WO2td9iNH35m7_lscYzXp5vsa-jyBiBH-LphQCeXLwL_HssrM5Oe6KqX_wI5sApYw2XM_5jOr_AGVO6VefE5oETJn35-OwJOxuPTg-Ps9RUIXNC59R6XupGV1ZZ6mYgQlk3eXAOvRxl0LxViPescqoJ2iK2UQi_XASRspIyOFs9ZWtt1_rnjPvCWVDB6lA70YAF14QiFB5hAOgqlAP2bqFFc9VzZxiMOUjjJmrckMZN1PiAHZCal_OI9Tr-0N38NGkTGWk9CJUPIZRaSBgCSKuEBxR66BtRDNjuYpFM2or4H0vDGbDXy2HcRJQZgdZ3tzgHYWtFbbxQ5mf9mi4lQcRZIebDEbWy2iuiro6004tI1I3RNcaDxfb_xXrFHqAlm88fJp922ENSSSyorHfZ2vzm1r9A0DO3L5M1_wHumQMa |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Efficient+Processing+of+Spatio-Temporal+Data+Streams+With+Spiking+Neural+Networks&rft.jtitle=Frontiers+in+neuroscience&rft.au=Kugele%2C+Alexander&rft.au=Pfeil%2C+Thomas&rft.au=Pfeiffer%2C+Michael&rft.au=Chicca%2C+Elisabetta&rft.date=2020-05-05&rft.pub=Frontiers+Media+S.A&rft.issn=1662-4548&rft.eissn=1662-453X&rft.volume=14&rft_id=info:doi/10.3389%2Ffnins.2020.00439&rft_id=info%3Apmid%2F32431592&rft.externalDBID=PMC7214871 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1662-4548&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1662-4548&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1662-4548&client=summon |