Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware
Abstract Mixed signal analog/digital neuromorphic circuits represent an ideal medium for reproducing bio-physically realistic dynamics of biological neural systems in real-time. However, similar to their biological counterparts, these circuits have limited resolution and are affected by a high degre...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-55749-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559324036431872 |
|---|---|
| author | Valentina Baruzzi Giacomo Indiveri Silvio P. Sabatini |
| author_facet | Valentina Baruzzi Giacomo Indiveri Silvio P. Sabatini |
| author_sort | Valentina Baruzzi |
| collection | DOAJ |
| description | Abstract Mixed signal analog/digital neuromorphic circuits represent an ideal medium for reproducing bio-physically realistic dynamics of biological neural systems in real-time. However, similar to their biological counterparts, these circuits have limited resolution and are affected by a high degree of variability. By developing a recurrent spiking neural network model of the retinocortical visual pathway, we show how such noisy and heterogeneous computing substrate can produce linear receptive fields tuned to visual stimuli with specific orientations and spatial frequencies. Compared to strictly feed-forward schemes, the model generates highly structured Gabor-like receptive fields of any phase symmetry, making optimal use of the hardware resources available in terms of synaptic connections and neuron numbers. Experimental results validate the approach, demonstrating how principles of neural computation can lead to robust sensory processing electronic systems, even when they are affected by high degree of heterogeneity, e.g., due to the use of analog circuits or memristive devices. |
| format | Article |
| id | doaj-art-e5dc0a3207cc4c9ba3552310cd89836b |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-e5dc0a3207cc4c9ba3552310cd89836b2025-01-05T12:38:11ZengNature PortfolioNature Communications2041-17232025-01-0116111510.1038/s41467-024-55749-yRecurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardwareValentina Baruzzi0Giacomo Indiveri1Silvio P. Sabatini2Department of Informatics, Bioengineering, Robotics and Systems Engineering, University of GenoaInstitute of Neuroinformatics, University of Zurich and ETH ZurichDepartment of Informatics, Bioengineering, Robotics and Systems Engineering, University of GenoaAbstract Mixed signal analog/digital neuromorphic circuits represent an ideal medium for reproducing bio-physically realistic dynamics of biological neural systems in real-time. However, similar to their biological counterparts, these circuits have limited resolution and are affected by a high degree of variability. By developing a recurrent spiking neural network model of the retinocortical visual pathway, we show how such noisy and heterogeneous computing substrate can produce linear receptive fields tuned to visual stimuli with specific orientations and spatial frequencies. Compared to strictly feed-forward schemes, the model generates highly structured Gabor-like receptive fields of any phase symmetry, making optimal use of the hardware resources available in terms of synaptic connections and neuron numbers. Experimental results validate the approach, demonstrating how principles of neural computation can lead to robust sensory processing electronic systems, even when they are affected by high degree of heterogeneity, e.g., due to the use of analog circuits or memristive devices.https://doi.org/10.1038/s41467-024-55749-y |
| spellingShingle | Valentina Baruzzi Giacomo Indiveri Silvio P. Sabatini Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware Nature Communications |
| title | Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware |
| title_full | Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware |
| title_fullStr | Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware |
| title_full_unstemmed | Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware |
| title_short | Recurrent models of orientation selectivity enable robust early-vision processing in mixed-signal neuromorphic hardware |
| title_sort | recurrent models of orientation selectivity enable robust early vision processing in mixed signal neuromorphic hardware |
| url | https://doi.org/10.1038/s41467-024-55749-y |
| work_keys_str_mv | AT valentinabaruzzi recurrentmodelsoforientationselectivityenablerobustearlyvisionprocessinginmixedsignalneuromorphichardware AT giacomoindiveri recurrentmodelsoforientationselectivityenablerobustearlyvisionprocessinginmixedsignalneuromorphichardware AT silviopsabatini recurrentmodelsoforientationselectivityenablerobustearlyvisionprocessinginmixedsignalneuromorphichardware |