File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Memristor-based analogue computing for brain-inspired sound localization with in situ training

TitleMemristor-based analogue computing for brain-inspired sound localization with in situ training
Authors
Issue Date2022
Citation
Nature Communications, 2022, v. 13, n. 1, article no. 2026 How to Cite?
AbstractThe human nervous system senses the physical world in an analogue but efficient way. As a crucial ability of the human brain, sound localization is a representative analogue computing task and often employed in virtual auditory systems. Different from well-demonstrated classification applications, all output neurons in localization tasks contribute to the predicted direction, introducing much higher challenges for hardware demonstration with memristor arrays. In this work, with the proposed multi-threshold-update scheme, we experimentally demonstrate the in-situ learning ability of the sound localization function in a 1K analogue memristor array. The experimental and evaluation results reveal that the scheme improves the training accuracy by ∼45.7% compared to the existing method and reduces the energy consumption by ∼184× relative to the previous work. This work represents a significant advance towards memristor-based auditory localization system with low energy consumption and high performance.
Persistent Identifierhttp://hdl.handle.net/10722/334827
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorGao, Bin-
dc.contributor.authorZhou, Ying-
dc.contributor.authorZhang, Qingtian-
dc.contributor.authorZhang, Shuanglin-
dc.contributor.authorYao, Peng-
dc.contributor.authorXi, Yue-
dc.contributor.authorLiu, Qi-
dc.contributor.authorZhao, Meiran-
dc.contributor.authorZhang, Wenqiang-
dc.contributor.authorLiu, Zhengwu-
dc.contributor.authorLi, Xinyi-
dc.contributor.authorTang, Jianshi-
dc.contributor.authorQian, He-
dc.contributor.authorWu, Huaqiang-
dc.date.accessioned2023-10-20T06:51:02Z-
dc.date.available2023-10-20T06:51:02Z-
dc.date.issued2022-
dc.identifier.citationNature Communications, 2022, v. 13, n. 1, article no. 2026-
dc.identifier.urihttp://hdl.handle.net/10722/334827-
dc.description.abstractThe human nervous system senses the physical world in an analogue but efficient way. As a crucial ability of the human brain, sound localization is a representative analogue computing task and often employed in virtual auditory systems. Different from well-demonstrated classification applications, all output neurons in localization tasks contribute to the predicted direction, introducing much higher challenges for hardware demonstration with memristor arrays. In this work, with the proposed multi-threshold-update scheme, we experimentally demonstrate the in-situ learning ability of the sound localization function in a 1K analogue memristor array. The experimental and evaluation results reveal that the scheme improves the training accuracy by ∼45.7% compared to the existing method and reduces the energy consumption by ∼184× relative to the previous work. This work represents a significant advance towards memristor-based auditory localization system with low energy consumption and high performance.-
dc.languageeng-
dc.relation.ispartofNature Communications-
dc.titleMemristor-based analogue computing for brain-inspired sound localization with in situ training-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1038/s41467-022-29712-8-
dc.identifier.pmid35440127-
dc.identifier.scopuseid_2-s2.0-85128369229-
dc.identifier.volume13-
dc.identifier.issue1-
dc.identifier.spagearticle no. 2026-
dc.identifier.epagearticle no. 2026-
dc.identifier.eissn2041-1723-
dc.identifier.isiWOS:000784997300102-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats