Unsupervised learning of visual features through spike timing dependent plasticity. Masquelier, T. & Thorpe, S. J PLoS Comput Biol, 3(2):e31, 2007. Place: United States ISBN: 1553-7358doi abstract bibtex Spike timing dependent plasticity (STDP) is a learning rule that modifies synaptic strength as a function of the relative timing of pre- and postsynaptic spikes. When a neuron is repeatedly presented with similar inputs, STDP is known to have the effect of concentrating high synaptic weights on afferents that systematically fire early, while postsynaptic spike latencies decrease. Here we use this learning rule in an asynchronous feedforward spiking neural network that mimics the ventral visual pathway and shows that when the network is presented with natural images, selectivity to intermediate-complexity visual features emerges. Those features, which correspond to prototypical patterns that are both salient and consistently present in the images, are highly informative and enable robust object recognition, as demonstrated on various classification tasks. Taken together, these results show that temporal codes may be a key to understanding the phenomenal processing speed achieved by the visual system and that STDP can lead to fast and selective responses.
@article{masquelier_unsupervised_2007,
title = {Unsupervised learning of visual features through spike timing dependent plasticity.},
volume = {3},
doi = {10.1371/journal.pcbi.0030031},
abstract = {Spike timing dependent plasticity (STDP) is a learning rule that modifies synaptic strength as a function of the relative timing of pre- and postsynaptic spikes. When a neuron is repeatedly presented with similar inputs, STDP is known to have the effect of concentrating high synaptic weights on afferents that systematically fire early, while postsynaptic spike latencies decrease. Here we use this learning rule in an asynchronous feedforward spiking neural network that mimics the ventral visual pathway and shows that when the network is presented with natural images, selectivity to intermediate-complexity visual features emerges. Those features, which correspond to prototypical patterns that are both salient and consistently present in the images, are highly informative and enable robust object recognition, as demonstrated on various classification tasks. Taken together, these results show that temporal codes may be a key to understanding the phenomenal processing speed achieved by the visual system and that STDP can lead to fast and selective responses.},
language = {eng},
number = {2},
journal = {PLoS Comput Biol},
author = {Masquelier, Timothée and Thorpe, Simon J},
year = {2007},
pmid = {17305422},
note = {Place: United States
ISBN: 1553-7358},
keywords = {Action Potentials, Artificial Intelligence, Computer Simulation, Feedback, Models, Neurological, Nerve Net, Neuronal Plasticity, Neurons, Afferent, Pattern Recognition, Visual, Synaptic Transmission, Visual Cortex, research support, non-u.s. gov't},
pages = {e31},
}
Downloads: 0
{"_id":"73cn43nhM8moAhusR","bibbaseid":"masquelier-thorpe-unsupervisedlearningofvisualfeaturesthroughspiketimingdependentplasticity-2007","downloads":0,"creationDate":"2016-03-16T14:38:25.750Z","title":"Unsupervised learning of visual features through spike timing dependent plasticity.","author_short":["Masquelier, T.","Thorpe, S. J"],"year":2007,"bibtype":"article","biburl":"https://bibbase.org/zotero/michaeltarr","bibdata":{"bibtype":"article","type":"article","title":"Unsupervised learning of visual features through spike timing dependent plasticity.","volume":"3","doi":"10.1371/journal.pcbi.0030031","abstract":"Spike timing dependent plasticity (STDP) is a learning rule that modifies synaptic strength as a function of the relative timing of pre- and postsynaptic spikes. When a neuron is repeatedly presented with similar inputs, STDP is known to have the effect of concentrating high synaptic weights on afferents that systematically fire early, while postsynaptic spike latencies decrease. Here we use this learning rule in an asynchronous feedforward spiking neural network that mimics the ventral visual pathway and shows that when the network is presented with natural images, selectivity to intermediate-complexity visual features emerges. Those features, which correspond to prototypical patterns that are both salient and consistently present in the images, are highly informative and enable robust object recognition, as demonstrated on various classification tasks. Taken together, these results show that temporal codes may be a key to understanding the phenomenal processing speed achieved by the visual system and that STDP can lead to fast and selective responses.","language":"eng","number":"2","journal":"PLoS Comput Biol","author":[{"propositions":[],"lastnames":["Masquelier"],"firstnames":["Timothée"],"suffixes":[]},{"propositions":[],"lastnames":["Thorpe"],"firstnames":["Simon","J"],"suffixes":[]}],"year":"2007","pmid":"17305422","note":"Place: United States ISBN: 1553-7358","keywords":"Action Potentials, Artificial Intelligence, Computer Simulation, Feedback, Models, Neurological, Nerve Net, Neuronal Plasticity, Neurons, Afferent, Pattern Recognition, Visual, Synaptic Transmission, Visual Cortex, research support, non-u.s. gov't","pages":"e31","bibtex":"@article{masquelier_unsupervised_2007,\n\ttitle = {Unsupervised learning of visual features through spike timing dependent plasticity.},\n\tvolume = {3},\n\tdoi = {10.1371/journal.pcbi.0030031},\n\tabstract = {Spike timing dependent plasticity (STDP) is a learning rule that modifies synaptic strength as a function of the relative timing of pre- and postsynaptic spikes. When a neuron is repeatedly presented with similar inputs, STDP is known to have the effect of concentrating high synaptic weights on afferents that systematically fire early, while postsynaptic spike latencies decrease. Here we use this learning rule in an asynchronous feedforward spiking neural network that mimics the ventral visual pathway and shows that when the network is presented with natural images, selectivity to intermediate-complexity visual features emerges. Those features, which correspond to prototypical patterns that are both salient and consistently present in the images, are highly informative and enable robust object recognition, as demonstrated on various classification tasks. Taken together, these results show that temporal codes may be a key to understanding the phenomenal processing speed achieved by the visual system and that STDP can lead to fast and selective responses.},\n\tlanguage = {eng},\n\tnumber = {2},\n\tjournal = {PLoS Comput Biol},\n\tauthor = {Masquelier, Timothée and Thorpe, Simon J},\n\tyear = {2007},\n\tpmid = {17305422},\n\tnote = {Place: United States\nISBN: 1553-7358},\n\tkeywords = {Action Potentials, Artificial Intelligence, Computer Simulation, Feedback, Models, Neurological, Nerve Net, Neuronal Plasticity, Neurons, Afferent, Pattern Recognition, Visual, Synaptic Transmission, Visual Cortex, research support, non-u.s. gov't},\n\tpages = {e31},\n}\n\n","author_short":["Masquelier, T.","Thorpe, S. J"],"key":"masquelier_unsupervised_2007","id":"masquelier_unsupervised_2007","bibbaseid":"masquelier-thorpe-unsupervisedlearningofvisualfeaturesthroughspiketimingdependentplasticity-2007","role":"author","urls":{},"keyword":["Action Potentials","Artificial Intelligence","Computer Simulation","Feedback","Models","Neurological","Nerve Net","Neuronal Plasticity","Neurons","Afferent","Pattern Recognition","Visual","Synaptic Transmission","Visual Cortex","research support","non-u.s. gov't"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"search_terms":["unsupervised","learning","visual","features","through","spike","timing","dependent","plasticity","masquelier","thorpe"],"keywords":["action potentials","artificial intelligence","computer simulation","feedback","models","neurological","nerve net","neuronal plasticity","neurons","afferent","pattern recognition","visual","synaptic transmission","visual cortex","research support","non-u.s. gov't"],"authorIDs":[],"dataSources":["v98BPm7u4WMoReKPh","5qXSH7BrePnXHtcrf"]}