Generalization in Mind and Machine

What do adversarial images tell us about human vision?

Autores: Marin Dujmović, Gaurav Malhotra, Jeffrey S Bowers
Publicado en: eLife, Edición 9, 2020, Página(s) 1-29, ISSN 2050-084X
Editor: eLife Sciences Publications
DOI: 10.7554/elife.55978

Can deep convolutional neural networks support relational reasoning in the same-different task?

Autores: Guillermo Puebla; Jeffrey S. Bowers
Publicado en: Journal of Vision, Edición 22 (10) 11, 2022, Página(s) 1-18, ISSN 1534-7362
Editor: Association for Research in Vision and Ophthalmology
DOI: 10.1167/jov.22.10.11

The role of capacity constraints in Convolutional Neural Networks for learning random versus natural data

Autores: Chris I Tsvetkov, Gaurav Malhotra, Benjamin D Evans, Jeffrey S Bowers
Publicado en: Neural Networks, Edición 161, 2023, Página(s) 515-524, ISSN 0346-251X
Editor: Pergamon Press Ltd.
DOI: 10.1016/j.neunet.2023.01.011

Convolutional Neural Networks Are Not Invariant to Translation, but They Can Learn to Be

Autores: Valerio Biscione; Jeffrey S Bowers
Publicado en: Journal of Machine Learning Research, 2021, Página(s) 1-28, ISSN 1532-4435
Editor: MIT Press
DOI: 10.48550/arxiv.2110.05861

Parallel Distributed Processing Theory in the Age of Deep Networks

Autores: Jeffrey S. Bowers
Publicado en: Trends in Cognitive Sciences, Edición 21/12, 2017, Página(s) 950-961, ISSN 1364-6613
Editor: Elsevier BV
DOI: 10.1016/j.tics.2017.09.013

Feature Blindness: A Challenge for Understanding and Modelling Visual Object Recognition

Autores: Gaurav Malhotra; Marin Dujmovic; Jeffrey S Bowers
Publicado en: PLOS Computational Biology, 2022, ISSN 1932-6203
Editor: Public Library of Science
DOI: 10.1101/2021.10.20.465074

Biological convolutions improve DNN robustness to noise and generalisation

Autores: Benjamin D. Evans; Gaurav Malhotra; Jeffrey S. Bowers
Publicado en: Neural Networks, Edición 3, 2022, Página(s) 96-110, ISSN 0893-6080
Editor: Pergamon Press Ltd.
DOI: 10.1016/j.neunet.2021.12.005

Successes and critical failures of neural networks in capturing human-like speech recognition

Autores: Federico Adolfi, Jeffrey S. Bowers, David Poeppel
Publicado en: Neural Networks, 2023, Página(s) 199-211, ISSN 0893-6080
Editor: Pergamon Press Ltd.
DOI: 10.1016/j.neunet.2023.02.032

Training neural networks to encode symbols enables combinatorial generalization

Autores: Ivan I. Vankov, Jeffrey S. Bowers
Publicado en: Philosophical Transactions of the Royal Society B: Biological Sciences, Edición 375/1791, 2019, Página(s) 20190309, ISSN 0962-8436
Editor: Royal Society of London
DOI: 10.1098/rstb.2019.0309

Hiding a plane with a pixel: examining shape-bias in CNNs and the benefit of building in biological constraints

Autores: Gaurav Malhotra, Benjamin D. Evans, Jeffrey S. Bowers
Publicado en: Vision Research, Edición 174, 2020, Página(s) 57-68, ISSN 0042-6989
Editor: Pergamon Press Ltd.
DOI: 10.1016/j.visres.2020.04.013

Does that sound right? A Novel Method of Evaluating Models of Reading Aloud.

Autores: Michele Gubian, Ryan Blything, Colin Davis, Jeffrey Bowers
Publicado en: Behavior Research Methods, Edición 2022, 2022, ISSN 1554-351X
Editor: Springer Verlag
DOI: 10.3758/s13428-022-01794-8

Are there any ‘object detectors’ in the hidden layers of CNNs trained to identify objects or scenes?

Autores: Ella M. Gale, Nicholas Martin, Ryan Blything, Anh Nguyen, Jeffrey S. Bowers
Publicado en: Vision Research, Edición 176, 2020, Página(s) 60-71, ISSN 0042-6989
Editor: Pergamon Press Ltd.
DOI: 10.1016/j.visres.2020.06.007

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

Autores: Ryan Blything, Valerio Biscione, Ivan I. Vankov, Casimir J. H. Ludwig, Jeffrey S. Bowers
Publicado en: Journal of Vision, Edición 21/2, 2021, Página(s) 9, ISSN 1534-7362
Editor: Association for Research in Vision and Ophthalmology
DOI: 10.1167/jov.21.2.9

The Pitfalls of Measuring Representational Similarity Using Representational Similarity Analysis

Autores: Marin Dujmović, Jeffrey S. Bowers, Federico Adolfi, Gaurav Malhotra
Publicado en: BioRxiv, 2022
Editor: Cold Spring Harbor Laboratory
DOI: 10.1101/2022.04.05.487135

Generalisation in Neural Networks Does not Require Feature Overlap

Autores: Jeff Mitchell; Jeffrey S. Bowers
Publicado en: ArXiv, 2021
Editor: Cornell University
DOI: 10.48550/arxiv.2107.06872

Learning Online Visual Invariances for Novel Objects via Supervised and Self-Supervised Training

Autores: Valerio Biscione; Jeffrey S. Bowers
Publicado en: ArXiv, 2021
Editor: Cornell University
DOI: 10.48550/arxiv.2110.01476

Do DNNs Trained on Natural Images Organize Visual Features into Gestalts?

Autores: Valerio Biscione, Jeffrey S. Bowers
Publicado en: Arxiv, 2022
Editor: Arxiv
DOI: 10.48550/arxiv.2203.07302

Deep Problems with Neural Network Models of Human Vision

Autores: Jeffrey S. Bowers, Gaurav Malhotra, Marin Dujmović, Milton Llera Montero, Christian Tsvetkov, Biscione, Guillermo Puebla, Federico Adolfi, John E. Hummel, Rachel F. Heaton, Benjamin D. Evans, Jeffrey Mitchell, Ryan Blything
Publicado en: PsyArXiv, 2022
Editor: Cornell University
DOI: 10.31234/osf.io/5zf4s

Can deep convolutional neural networks support relational reasoning in the same-different task?

Autores: Guillermo Puebla; Jeffrey S. Bowers
Publicado en: BioRxiv, 2021
Editor: Cold Spring Harbor Laboratory
DOI: 10.1101/2021.09.03.458919

The contrasting shape representations that support object recognition in humans and CNNs

Autores: Gaurav Malhotra; Marin Dujmovic; John Hummel; Jeffrey S Bowers
Publicado en: BioRxiv, 2021
Editor: Cold Spring Harbor Laboratory
DOI: 10.1101/2021.12.14.472546

Lost in Latent Space: Disentangled Models and the Challenge of Combinatorial Generalisation

Autores: Milton Montero, Jeffrey S. Bowers, Rui Ponte Costa, Casimir J. Ludwig, Gaurav Malhotra
Publicado en: arXiv, 2022
Editor: Cornell Univesity
DOI: 10.48550/arxiv.2204.02283

A case for robust translation tolerance in humans and CNNs. A commentary on Han et al

Autores: Blything, Ryan; Biscione, Valerio; Bowers, Jeffrey
Publicado en: arXiv.org e-Print Archive, Edición 1, 2020
Editor: arXiv
DOI: 10.48550/arxiv.2012.05950

Learning Translation Invariance in CNNs

Autores: Valerio Biscione, Jeffrey Bowers
Publicado en: NeurIPS 2020 Workshop SVRHM, 2020
Editor: NeurIPS 2020 Workshop SVRHM

The role of disentanglement in generalisation

Autores: Milton Llera Montero; Casimir Ludwig; Rui Ponte Costa; Gaurav Malhotra; Jeffrey S Bowers
Publicado en: International Conference on Learning Representations, 2021
Editor: OpenReview

Can Deep Convolutional Neural Networks Learn Same-Different Relations?

Autores: Guillermo Puebla; Jeffrey S. Bowers
Publicado en: Proceedings of the Annual Meeting of the Cognitive Science Society, 43(43), Edición 43(43), 2021, ISSN 1069-7977
Editor: University of California

Subtractive gating improves generalization in working memory tasks

Autores: Milton Llera Montero, Gaurav Malhotra, Jeff Bowers, Rui Ponte Costa
Publicado en: 2019 Conference on Cognitive Computational Neuroscience, 2019
Editor: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1352-0

Do LSTMs know about Principle C?

Autores: Jeff Mitchell, Nina Kazanina, Conor Houghton, Jeff Bowers
Publicado en: 2019 Conference on Cognitive Computational Neuroscience, 2019
Editor: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1241-0

Extreme Translation Tolerance in Humans and Machines

Autores: Ryan Blything, Ivan Vankov, Casimir Ludwig, Jeffrey Bowers
Publicado en: 2019 Conference on Cognitive Computational Neuroscience, 2019
Editor: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1091-0

Humans cannot decipher adversarial images: Revisiting Zhou and Firestone (2019)

Autores: Marin Dujmović, Gaurav Malhotra, Jeffrey Bowers
Publicado en: 2019 Conference on Cognitive Computational Neuroscience, 2019
Editor: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1298-0

Adding biological constraints to CNNs makes image classification more human-like and robust

Autores: Gaurav Malhotra, Benjamin Evans, Jeffrey Bowers
Publicado en: 2019 Conference on Cognitive Computational Neuroscience, 2019
Editor: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1212-0

Translation Tolerance in Vision.

Autores: Ryan Blything, Ivan I. Vankov, Casimir J. Ludwig, Jeffrey S. Bowers
Publicado en: 41st Annual Conference of the Cognitive Science Society 2019, 2019
Editor: A.K. Goel, C.M. Seifert, & C. Freksa (Eds.)

Selectivity Metrics Provide Misleading Estimates of the Selectivity of Single Units in Neural Networks

Autores: Ella Gale, Ryan Blything, Nicholas Martin, Jeffrey S. Bowers, and Anh Nguyen.
Publicado en: 41st Annual Conference of the Cognitive Science Society 2019, 2019
Editor: A.K. Goel, C.M. Seifert, & C. Freksa (Eds.)

The Contrasting Roles of Shape in Human Vision and Convolutional Neural Networks

Autores: Gaurav Malhotra, Jeff Bowers
Publicado en: 41st Annual Conference of the Cognitive Science Society 2019, 2019
Editor: A.K. Goel, C.M. Seifert, & C. Freksa (Eds.)

Harnessing the Symmetry of Convolutions for Systematic Generalisation

Autores: Jeff Mitchell, Jeffrey S. Bowers
Publicado en: 2020 International Joint Conference on Neural Networks (IJCNN), 2020, Página(s) 1-8, ISBN 978-1-7281-6926-2
Editor: IEEE
DOI: 10.1109/ijcnn48605.2020.9207183

Adding biological constraints to deep neural networks reduces their capacity to learn unstructured data

Autores: Christian Tsvetkov, Gaurav Malhotra, Benjamin Evans, Jeff Bowers
Publicado en: In Proceedings of the 42nd Annual Conference of the Cognitive Science Society 2020, 2020, Página(s) 2358-2364
Editor: Cognitive Science Society

Priorless Recurrent Networks Learn Curiously

Autores: Jeff Mitchell, Jeffrey Bowers
Publicado en: Proceedings of the 28th International Conference on Computational Linguistics, 2020, Página(s) 5147-5158, ISBN 978-1-952148-27-9
Editor: International Committee on Computational Linguistics
DOI: 10.18653/v1/2020.coling-main.451

Publicaciones

Compartir esta página

Descargar