The human visual system and CNNs can both support robust online translation tolerance following extreme displacements

Autor: Ivan Vankov, Jeffrey S. Bowers, Valerio Biscione, Ryan Blything, Casimir J. H. Ludwig

Publikováno v: Explore Bristol Research
arXiv.org e-Print Archive
PubMed Central
CORE (RIOXX-UK Aggregator)
ORCID
Microsoft Academic Graph
Journal of Vision
Blything, R, Biscione, V, Vankov, I, Ludwig, C J H & Bowers, J S 2021, ' The human visual system and CNNs can both support robust online translation tolerance following extreme displacements ', Journal of Vision, vol. 21, no. 2, 9, pp. 1-16 . https://doi.org/10.1167/jov.21.2.9

Comment: Main manuscript contains 5 figures plus 2 tables. SI contains 2 tables
Visual translation tolerance refers to our capacity to recognize objects over a wide range of different retinal locations. Although translation is perhaps the simple

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::86d9f6381dd6bb974016b0bfa32eaade
http://arxiv.org/abs/2009.12855

Do arbitrary input–output mappings in parallel distributed processing networks require localist coding?

Autor: Ivan Vankov, Jeffrey S. Bowers

Publikováno v: Vankov, I & Bowers, J 2017, ' Do arbitrary input-output mappings in parallel distributed processing networks require localist coding? ', Language, Cognition and Neuroscience, vol. 32, no. 3, pp. 392-399 . https://doi.org/10.1080/23273798.2016.1256490

The Parallel Distributed Processing (PDP) approach to cognitive modeling assumes that knowledge is distributed across multiple processing units. This view is typically justified on the basis of the computational advantages and biological plausibility

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::909b249c580162ebf467e881a284108f
https://doi.org/10.1080/23273798.2016.1256490

Training neural networks to encode symbols enables combinatorial generalization

Autor: Ivan Vankov, Jeffrey S. Bowers

Publikováno v: Philos Trans R Soc Lond B Biol Sci
Sygma
Explore Bristol Research
Datacite
UnpayWall
ORCID
Microsoft Academic Graph
arXiv.org e-Print Archive
European Union Open Data Portal
PubMed Central
Philosophical Transactions of the Royal Society B: Biological Sciences
Vankov, I I & Bowers, J S 2019, ' Training neural networks to encode symbols enables combinatorial generalization ', Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 375, no. 1791, 20190309 . https://doi.org/10.1098/rstb.2019.0309

Combinatorial generalization—the ability to understand and produce novel combinations of already familiar elements—is considered to be a core capacity of the human mind and a major challenge to neural network models. A significant body of researc

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bd821c7ba5ae1a07a01835da0d994920
https://pubmed.ncbi.nlm.nih.gov/31840580

Manipulating the alpha level cannot cure significance testing – comments on 'Redefine statistical significance'

Autor: Armina Janyan, Marion Tegethoff, Martin E. Guerrero-Gimenez, Rubén Daniel Ledesma, Fränzi Korner-Nievergelt, James W. Grice, Yuki Yamada, Corson N. Areshenkoff, Carlos Barrera-Causil, Michiel R. de Boer, Michael J. Marks, Ivan Vankov, Felipe Carlos Martín Zoppino, Martin Lachmair, Susana Ruiz-Fernández, Koji Kosugi, Isabel Suarez, Marian Grendar, Tonghui Wang, Juan José Rahona, Subhra Sankar Dhar, Valentin Amrhein, Jose D Perezgonzalez, Ladislas Nalborczyk, David A. Rodriguez-Medina, Eric J. Beh, Juan Carlos Correa, M. T. Bradley, Fernando Marmolejo-Ramos, William M. Briggs, Rens van de Schoot, Marco Tullio Liuzza, Daniel R. Ciocca, Roberto Limongi, Rosaria Lombardo, Juana Gómez-Benito, Igor Dolgov, Andrés Gutiérrez, Roland Pfister, Yusuf K. Bilgic, Héctor A. Cepeda-Freyre, Ali Karimnezhad, Tania B. Huedo-Medina, David Trafimow, Gunther Meinlschmidt, Denis Cousineau, Sergio E. Chaigneau, Raydonal Ospina, Roser Bono, Xavier Romão, Klaus Jaffe, Santiago Velasco-Forero, Mauricio Tejo, Hung T. Nguyen

We argue that depending on p-values to reject null hypotheses, including a recent call for changing the canonical alpha level for statistical significance from .05 to .005, is deleterious for the finding of new discoveries and the progress of science

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c40b70ac564b467eb0aec12fcd9164a

Neural networks learn highly selective representations in order to overcome the superposition catastrophe

Autor: Colin J. Davis, Jeffrey S. Bowers, Markus F. Damian, Ivan Vankov

Publikováno v: Psychological Review. 121:248-261

A key insight from 50 years of neurophysiology is that some neurons in cortex respond to information in a highly selective manner. Why is this? We argue that selective representations support the coactivation of multiple "things" (e.g., words, object

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ea66f04cadc744aa5ab37ce9b5e219a0
https://doi.org/10.1037/a0035943