Visual Cortex and Deep Networks
A mathematical framework that describes learning of invariant representations in the ventral stream, offering both theoretical development and applications.The ventral visual stream is believed to underlie object recognition in primates. Over the past fifty years, researchers have developed a series...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Online |
| Language: | English |
| Published: |
The MIT Press
2024
|
| Subjects: | |
| Online Access: | ONIX_20241025_9780262336710_14 |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1869518526872551424 |
|---|---|
| author | Poggio, Tomaso A. Anselmi, Fabio |
| author_browse | Anselmi, Fabio Poggio, Tomaso A. |
| author_facet | Poggio, Tomaso A. Anselmi, Fabio |
| author_sort | Poggio, Tomaso A. |
| collection | Directory of Open Access Books |
| description | A mathematical framework that describes learning of invariant representations in the ventral stream, offering both theoretical development and applications.The ventral visual stream is believed to underlie object recognition in primates. Over the past fifty years, researchers have developed a series of quantitative models that are increasingly faithful to the biological architecture. Recently, deep learning convolution networks—which do not reflect several important features of the ventral stream architecture and physiology—have been trained with extremely large datasets, resulting in model neurons that mimic object recognition but do not explain the nature of the computations carried out in the ventral stream. This book develops a mathematical framework that describes learning of invariant representations of the ventral stream and is particularly relevant to deep convolutional learning networks. The authors propose a theory based on the hypothesis that the main computational goal of the ventral stream is to compute neural representations of images that are invariant to transformations commonly encountered in the visual environment and are learned from unsupervised experience. They describe a general theoretical framework of a computational theory of invariance (with details and proofs offered in appendixes) and then review the application of the theory to the feedforward path of the ventral stream in the primate visual cortex. |
| format | Online |
| id | doab-20.500.12854ir-146636 |
| institution | Directory of Open Access Books |
| language | eng |
| publishDate | 2024 |
| publishDateRange | 2024 |
| publishDateSort | 2024 |
| publisher | The MIT Press |
| publisherStr | The MIT Press |
| record_format | ojs |
| spelling | doab-20.500.12854ir-1466362024-10-25T13:13:54Z Visual Cortex and Deep Networks Poggio, Tomaso A. Anselmi, Fabio visual cortex deep networks deep learning computational neuroscience invariant representations ventral stream computer vision feed forward thema EDItEUR::P Mathematics and Science::PS Biology, life sciences::PSA Life sciences: general issues::PSAN Neurosciences thema EDItEUR::U Computing and Information Technology::UY Computer science::UYQ Artificial intelligence::UYQM Machine learning A mathematical framework that describes learning of invariant representations in the ventral stream, offering both theoretical development and applications.The ventral visual stream is believed to underlie object recognition in primates. Over the past fifty years, researchers have developed a series of quantitative models that are increasingly faithful to the biological architecture. Recently, deep learning convolution networks—which do not reflect several important features of the ventral stream architecture and physiology—have been trained with extremely large datasets, resulting in model neurons that mimic object recognition but do not explain the nature of the computations carried out in the ventral stream. This book develops a mathematical framework that describes learning of invariant representations of the ventral stream and is particularly relevant to deep convolutional learning networks. The authors propose a theory based on the hypothesis that the main computational goal of the ventral stream is to compute neural representations of images that are invariant to transformations commonly encountered in the visual environment and are learned from unsupervised experience. They describe a general theoretical framework of a computational theory of invariance (with details and proofs offered in appendixes) and then review the application of the theory to the feedforward path of the ventral stream in the primate visual cortex. 2024-10-25T13:13:51Z 2024-10-25T13:13:51Z 2016 book ONIX_20241025_9780262336710_14 9780262336710 9780262034722 https://directory.doabooks.org/handle/20.500.12854/146636 eng Computational Neuroscience Series image/jpeg n/a https://doi.org/10.7551/mitpress/10177.001.0001 The MIT Press The MIT Press 10.7551/mitpress/10177.001.0001 10.7551/mitpress/10177.001.0001 ae0cf962-f685-4933-93d1-916defa5123d 9780262336710 9780262034722 The MIT Press 136 Cambridge open access |
| spellingShingle | visual cortex deep networks deep learning computational neuroscience invariant representations ventral stream computer vision feed forward thema EDItEUR::P Mathematics and Science::PS Biology, life sciences::PSA Life sciences: general issues::PSAN Neurosciences thema EDItEUR::U Computing and Information Technology::UY Computer science::UYQ Artificial intelligence::UYQM Machine learning Poggio, Tomaso A. Anselmi, Fabio Visual Cortex and Deep Networks |
| title | Visual Cortex and Deep Networks |
| title_full | Visual Cortex and Deep Networks |
| title_fullStr | Visual Cortex and Deep Networks |
| title_full_unstemmed | Visual Cortex and Deep Networks |
| title_short | Visual Cortex and Deep Networks |
| title_sort | visual cortex and deep networks |
| topic | visual cortex deep networks deep learning computational neuroscience invariant representations ventral stream computer vision feed forward thema EDItEUR::P Mathematics and Science::PS Biology, life sciences::PSA Life sciences: general issues::PSAN Neurosciences thema EDItEUR::U Computing and Information Technology::UY Computer science::UYQ Artificial intelligence::UYQM Machine learning |
| topic_facet | visual cortex deep networks deep learning computational neuroscience invariant representations ventral stream computer vision feed forward thema EDItEUR::P Mathematics and Science::PS Biology, life sciences::PSA Life sciences: general issues::PSAN Neurosciences thema EDItEUR::U Computing and Information Technology::UY Computer science::UYQ Artificial intelligence::UYQM Machine learning |
| url | ONIX_20241025_9780262336710_14 |
| work_keys_str_mv | AT poggiotomasoa visualcortexanddeepnetworks AT anselmifabio visualcortexanddeepnetworks |