Generic decoding of seen and imagined objects using hierarchical visual features

Generic decoding of seen and imagined objects using hierarchical visual features

Machine learning algorithms can decode objects that people see or imagine from their brain activity. Here the authors present a predictive decoder combined with deep neural network representations that generalizes beyond the training set and correctly identifies novel objects that it has never been...

Full description

Bibliographic Details
Main Authors: Tomoyasu Horikawa, Yukiyasu Kamitani
Format: Article
Language:English
Published: Nature Publishing Group 2017-05-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/ncomms15037
id doaj-6ccae50a9ded47ada2c491896e3c5bb9
record_format Article
spelling doaj-6ccae50a9ded47ada2c491896e3c5bb92021-05-11T07:42:02ZengNature Publishing GroupNature Communications2041-17232017-05-018111510.1038/ncomms15037Generic decoding of seen and imagined objects using hierarchical visual featuresTomoyasu Horikawa0Yukiyasu Kamitani1ATR Computational Neuroscience LaboratoriesATR Computational Neuroscience LaboratoriesMachine learning algorithms can decode objects that people see or imagine from their brain activity. Here the authors present a predictive decoder combined with deep neural network representations that generalizes beyond the training set and correctly identifies novel objects that it has never been trained on.https://doi.org/10.1038/ncomms15037
collection DOAJ
language English
format Article
sources DOAJ
author Tomoyasu Horikawa
Yukiyasu Kamitani
spellingShingle Tomoyasu Horikawa
Yukiyasu Kamitani
Generic decoding of seen and imagined objects using hierarchical visual features
Nature Communications
author_facet Tomoyasu Horikawa
Yukiyasu Kamitani
author_sort Tomoyasu Horikawa
title Generic decoding of seen and imagined objects using hierarchical visual features
title_short Generic decoding of seen and imagined objects using hierarchical visual features
title_full Generic decoding of seen and imagined objects using hierarchical visual features
title_fullStr Generic decoding of seen and imagined objects using hierarchical visual features
title_full_unstemmed Generic decoding of seen and imagined objects using hierarchical visual features
title_sort generic decoding of seen and imagined objects using hierarchical visual features
publisher Nature Publishing Group
series Nature Communications
issn 2041-1723
publishDate 2017-05-01
description Machine learning algorithms can decode objects that people see or imagine from their brain activity. Here the authors present a predictive decoder combined with deep neural network representations that generalizes beyond the training set and correctly identifies novel objects that it has never been trained on.
url https://doi.org/10.1038/ncomms15037
work_keys_str_mv AT tomoyasuhorikawa genericdecodingofseenandimaginedobjectsusinghierarchicalvisualfeatures
AT yukiyasukamitani genericdecodingofseenandimaginedobjectsusinghierarchicalvisualfeatures
_version_ 1721451767026155520

Similar Items