Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network
Synchronization has been suggested as a mechanism of binding distributed feature representations facilitating segmentation of visual stimuli. Here we investigate this concept based on unsupervised learning using natural visual stimuli. We simulate dual-variable neural oscillators with separate activ...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2014-01-01
|
| Series: | Frontiers in Computational Neuroscience |
| Subjects: | |
| Online Access: | http://journal.frontiersin.org/Journal/10.3389/fncom.2013.00195/full |
| _version_ | 1819230091758010368 |
|---|---|
| author | Holger eFinger Peter eKönig Peter eKönig |
| author_facet | Holger eFinger Peter eKönig Peter eKönig |
| author_sort | Holger eFinger |
| collection | DOAJ |
| description | Synchronization has been suggested as a mechanism of binding distributed feature representations facilitating segmentation of visual stimuli. Here we investigate this concept based on unsupervised learning using natural visual stimuli. We simulate dual-variable neural oscillators with separate activation and phase variables. The binding of a set of neurons is coded by synchronized phase variables. The network of tangential synchronizing connections learned from the induced activations exhibits small-world properties and allows binding even over larger distances. We evaluate the resulting dynamic phase maps using segmentation masks labeled by human experts. Our simulation results show a continuously increasing phase synchrony between neurons within the labeled segmentation masks. The evaluation of the network dynamics shows that the synchrony between network nodes establishes a relational coding of the natural image inputs. This demonstrates that the concept of binding by synchrony is applicable in the context of unsupervised learning using natural visual stimuli. |
| first_indexed | 2024-12-23T11:23:34Z |
| format | Article |
| id | doaj.art-254e5ff4df5c40ffa73c0d164f196ded |
| institution | Directory Open Access Journal |
| issn | 1662-5188 |
| language | English |
| last_indexed | 2024-12-23T11:23:34Z |
| publishDate | 2014-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Computational Neuroscience |
| spelling | doaj.art-254e5ff4df5c40ffa73c0d164f196ded2022-12-21T17:49:00ZengFrontiers Media S.A.Frontiers in Computational Neuroscience1662-51882014-01-01710.3389/fncom.2013.0019558951Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator networkHolger eFinger0Peter eKönig1Peter eKönig2University of OsnabrückUniversity of OsnabrückUniversity Medical Center Hamburg-EppendorfSynchronization has been suggested as a mechanism of binding distributed feature representations facilitating segmentation of visual stimuli. Here we investigate this concept based on unsupervised learning using natural visual stimuli. We simulate dual-variable neural oscillators with separate activation and phase variables. The binding of a set of neurons is coded by synchronized phase variables. The network of tangential synchronizing connections learned from the induced activations exhibits small-world properties and allows binding even over larger distances. We evaluate the resulting dynamic phase maps using segmentation masks labeled by human experts. Our simulation results show a continuously increasing phase synchrony between neurons within the labeled segmentation masks. The evaluation of the network dynamics shows that the synchrony between network nodes establishes a relational coding of the natural image inputs. This demonstrates that the concept of binding by synchrony is applicable in the context of unsupervised learning using natural visual stimuli.http://journal.frontiersin.org/Journal/10.3389/fncom.2013.00195/fullbindingsynchronizationoscillationnatural image statisticsunsupervised learningscene segmentation |
| spellingShingle | Holger eFinger Peter eKönig Peter eKönig Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network Frontiers in Computational Neuroscience binding synchronization oscillation natural image statistics unsupervised learning scene segmentation |
| title | Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network |
| title_full | Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network |
| title_fullStr | Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network |
| title_full_unstemmed | Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network |
| title_short | Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network |
| title_sort | phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network |
| topic | binding synchronization oscillation natural image statistics unsupervised learning scene segmentation |
| url | http://journal.frontiersin.org/Journal/10.3389/fncom.2013.00195/full |
| work_keys_str_mv | AT holgerefinger phasesynchronyfacilitatesbindingandsegmentationofnaturalimagesinacoupledneuraloscillatornetwork AT peterekonig phasesynchronyfacilitatesbindingandsegmentationofnaturalimagesinacoupledneuraloscillatornetwork AT peterekonig phasesynchronyfacilitatesbindingandsegmentationofnaturalimagesinacoupledneuraloscillatornetwork |