File Download
  Links for fulltext
     (May Require Subscription)

Article: Dropout in neural networks simulates the paradoxical effects of deep brain stimulation on memory

TitleDropout in neural networks simulates the paradoxical effects of deep brain stimulation on memory
Authors
KeywordsNeuromodulation
Deep brain stimulation
Memory
Neural network
Dropout
Issue Date2020
PublisherFrontiers Research Foundation. The Journal's web site is located at http://www.frontiersin.org/aging_neuroscience
Citation
Frontiers in Aging Neuroscience, 2020, v. 12, article no. 273 How to Cite?
AbstractNeuromodulation techniques such as Deep Brain Stimulation (DBS) are a promising treatment for memory-related disorders including anxiety, addiction, and dementia. However, the outcomes of such treatments appears to be somewhat paradoxical, in that these techniques can both disrupt and enhance memory even when applied to the same brain target. In this paper, we hypothesise that disruption and enhancement of memory through neuromodulation can be explained by the dropout of engram nodes. We used a convolutional neural network to classify handwritten digits and letters, and applied dropout at different stages to simulate DBS effects on engrams. We showed that dropout applied during training improved the accuracy of prediction, whereas dropout applied during testing dramatically decreased the accuracy of prediction, which mimics enhancement and disruption of memory, respectively. We further showed that transfer learning of neural networks with dropout had increased accuracy and rate of learning. Dropout during training provided a more robust “skeleton” network, and together with transfer learning, mimics the effects of chronic DBS on memory. Overall, we showed that the dropout of engram nodes is a possible mechanism by which neuromodulation techniques such as DBS can both disrupt and enhance memory, providing a unique perspective on this paradox.
Persistent Identifierhttp://hdl.handle.net/10722/285434
ISSN
2019 Impact Factor: 4.362
2015 SCImago Journal Rankings: 1.808

 

DC FieldValueLanguage
dc.contributor.authorTan, SZK-
dc.contributor.authorDu, R-
dc.contributor.authorPerucho, JAU-
dc.contributor.authorChopra, SS-
dc.contributor.authorVardhanabhuti, V-
dc.contributor.authorLim, LW-
dc.date.accessioned2020-08-18T03:53:22Z-
dc.date.available2020-08-18T03:53:22Z-
dc.date.issued2020-
dc.identifier.citationFrontiers in Aging Neuroscience, 2020, v. 12, article no. 273-
dc.identifier.issn1663-4365-
dc.identifier.urihttp://hdl.handle.net/10722/285434-
dc.description.abstractNeuromodulation techniques such as Deep Brain Stimulation (DBS) are a promising treatment for memory-related disorders including anxiety, addiction, and dementia. However, the outcomes of such treatments appears to be somewhat paradoxical, in that these techniques can both disrupt and enhance memory even when applied to the same brain target. In this paper, we hypothesise that disruption and enhancement of memory through neuromodulation can be explained by the dropout of engram nodes. We used a convolutional neural network to classify handwritten digits and letters, and applied dropout at different stages to simulate DBS effects on engrams. We showed that dropout applied during training improved the accuracy of prediction, whereas dropout applied during testing dramatically decreased the accuracy of prediction, which mimics enhancement and disruption of memory, respectively. We further showed that transfer learning of neural networks with dropout had increased accuracy and rate of learning. Dropout during training provided a more robust “skeleton” network, and together with transfer learning, mimics the effects of chronic DBS on memory. Overall, we showed that the dropout of engram nodes is a possible mechanism by which neuromodulation techniques such as DBS can both disrupt and enhance memory, providing a unique perspective on this paradox.-
dc.languageeng-
dc.publisherFrontiers Research Foundation. The Journal's web site is located at http://www.frontiersin.org/aging_neuroscience-
dc.relation.ispartofFrontiers in Aging Neuroscience-
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.subjectNeuromodulation-
dc.subjectDeep brain stimulation-
dc.subjectMemory-
dc.subjectNeural network-
dc.subjectDropout-
dc.titleDropout in neural networks simulates the paradoxical effects of deep brain stimulation on memory-
dc.typeArticle-
dc.identifier.emailVardhanabhuti, V: varv@hku.hk-
dc.identifier.emailLim, LW: limlw@hku.hk-
dc.identifier.authorityVardhanabhuti, V=rp01900-
dc.identifier.authorityLim, LW=rp02088-
dc.description.naturepublished_or_final_version-
dc.identifier.doi10.3389/fnagi.2020.00273-
dc.identifier.hkuros312833-
dc.identifier.volume12-
dc.identifier.spagearticle no. 273-
dc.identifier.epagearticle no. 273-
dc.publisher.placeSwitzerland-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats