Cumulative inhibition in neural networks
Tjøstheim, Trond Arild LU and Balkenius, Christian LU
(2019)
In Cognitive Processing
20(1).
p.87-102
- Abstract
- We show how a multi-resolution network can model the development of acuity and coarse-to-fine processing in the mammalian visual cortex. The network adapts to input statistics in an unsupervised manner, and learns a coarse-to-fine representation by using cumulative inhibition of nodes within a network layer. We show that a system of such layers can represent input by hierarchically composing larger parts from smaller components. It can also model aspects of top-down processes, such as image regeneration.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/93c5ca41-ce5b-4ea7-8a8e-d82533013fc1
- author
- Tjøstheim, Trond Arild
LU
and Balkenius, Christian
LU
- organization
- publishing date
- 2019
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Cumulative inhibition, Multi-resolution, Coarse-to-fine processing, Unsupervised learning, Acuity, Cortical microcolumn, Visual cortex
- in
- Cognitive Processing
- volume
- 20
- issue
- 1
- pages
- 87 - 102
- publisher
- Springer
- external identifiers
-
- pmid:30392141
- scopus:85055976065
- ISSN
- 1612-4782
- DOI
- 10.1007/s10339-018-0888-z
- project
- eSSENCE@LU 4:1 - Method development for analysis and modelling of large scale electrophysiological recordings using deep artificial neural networks
- Cognitive Philosophy Research Group (CogPhi)
- language
- English
- LU publication?
- yes
- id
- 93c5ca41-ce5b-4ea7-8a8e-d82533013fc1
- date added to LUP
- 2018-11-07 13:59:06
- date last changed
- 2022-05-03 07:26:31
@article{93c5ca41-ce5b-4ea7-8a8e-d82533013fc1,
abstract = {{We show how a multi-resolution network can model the development of acuity and coarse-to-fine processing in the mammalian visual cortex. The network adapts to input statistics in an unsupervised manner, and learns a coarse-to-fine representation by using cumulative inhibition of nodes within a network layer. We show that a system of such layers can represent input by hierarchically composing larger parts from smaller components. It can also model aspects of top-down processes, such as image regeneration.}},
author = {{Tjøstheim, Trond Arild and Balkenius, Christian}},
issn = {{1612-4782}},
keywords = {{Cumulative inhibition; Multi-resolution; Coarse-to-fine processing; Unsupervised learning; Acuity; Cortical microcolumn; Visual cortex}},
language = {{eng}},
number = {{1}},
pages = {{87--102}},
publisher = {{Springer}},
series = {{Cognitive Processing}},
title = {{Cumulative inhibition in neural networks}},
url = {{http://dx.doi.org/10.1007/s10339-018-0888-z}},
doi = {{10.1007/s10339-018-0888-z}},
volume = {{20}},
year = {{2019}},
}