Generalization in Mind and Machine

What do adversarial images tell us about human vision?

Autorzy: Marin Dujmović, Gaurav Malhotra, Jeffrey S Bowers
Opublikowane w: eLife, Numer 9, 2020, Strona(/y) 1-29, ISSN 2050-084X
Wydawca: eLife Sciences Publications
DOI: 10.7554/elife.55978

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

Autorzy: Guillermo Puebla; Jeffrey S. Bowers
Opublikowane w: Journal of Vision, Numer 22 (10) 11, 2022, Strona(/y) 1-18, ISSN 1534-7362
Wydawca: 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

Autorzy: Chris I Tsvetkov, Gaurav Malhotra, Benjamin D Evans, Jeffrey S Bowers
Opublikowane w: Neural Networks, Numer 161, 2023, Strona(/y) 515-524, ISSN 0346-251X
Wydawca: 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

Autorzy: Valerio Biscione; Jeffrey S Bowers
Opublikowane w: Journal of Machine Learning Research, 2021, Strona(/y) 1-28, ISSN 1532-4435
Wydawca: MIT Press
DOI: 10.48550/arxiv.2110.05861

Parallel Distributed Processing Theory in the Age of Deep Networks

Autorzy: Jeffrey S. Bowers
Opublikowane w: Trends in Cognitive Sciences, Numer 21/12, 2017, Strona(/y) 950-961, ISSN 1364-6613
Wydawca: Elsevier BV
DOI: 10.1016/j.tics.2017.09.013

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

Autorzy: Gaurav Malhotra; Marin Dujmovic; Jeffrey S Bowers
Opublikowane w: PLOS Computational Biology, 2022, ISSN 1932-6203
Wydawca: Public Library of Science
DOI: 10.1101/2021.10.20.465074

Biological convolutions improve DNN robustness to noise and generalisation

Autorzy: Benjamin D. Evans; Gaurav Malhotra; Jeffrey S. Bowers
Opublikowane w: Neural Networks, Numer 3, 2022, Strona(/y) 96-110, ISSN 0893-6080
Wydawca: Pergamon Press Ltd.
DOI: 10.1016/j.neunet.2021.12.005

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

Autorzy: Federico Adolfi, Jeffrey S. Bowers, David Poeppel
Opublikowane w: Neural Networks, 2023, Strona(/y) 199-211, ISSN 0893-6080
Wydawca: Pergamon Press Ltd.
DOI: 10.1016/j.neunet.2023.02.032

Training neural networks to encode symbols enables combinatorial generalization

Autorzy: Ivan I. Vankov, Jeffrey S. Bowers
Opublikowane w: Philosophical Transactions of the Royal Society B: Biological Sciences, Numer 375/1791, 2019, Strona(/y) 20190309, ISSN 0962-8436
Wydawca: 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

Autorzy: Gaurav Malhotra, Benjamin D. Evans, Jeffrey S. Bowers
Opublikowane w: Vision Research, Numer 174, 2020, Strona(/y) 57-68, ISSN 0042-6989
Wydawca: Pergamon Press Ltd.
DOI: 10.1016/j.visres.2020.04.013

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

Autorzy: Michele Gubian, Ryan Blything, Colin Davis, Jeffrey Bowers
Opublikowane w: Behavior Research Methods, Numer 2022, 2022, ISSN 1554-351X
Wydawca: 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?

Autorzy: Ella M. Gale, Nicholas Martin, Ryan Blything, Anh Nguyen, Jeffrey S. Bowers
Opublikowane w: Vision Research, Numer 176, 2020, Strona(/y) 60-71, ISSN 0042-6989
Wydawca: 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

Autorzy: Ryan Blything, Valerio Biscione, Ivan I. Vankov, Casimir J. H. Ludwig, Jeffrey S. Bowers
Opublikowane w: Journal of Vision, Numer 21/2, 2021, Strona(/y) 9, ISSN 1534-7362
Wydawca: Association for Research in Vision and Ophthalmology
DOI: 10.1167/jov.21.2.9

The Pitfalls of Measuring Representational Similarity Using Representational Similarity Analysis

Autorzy: Marin Dujmović, Jeffrey S. Bowers, Federico Adolfi, Gaurav Malhotra
Opublikowane w: BioRxiv, 2022
Wydawca: Cold Spring Harbor Laboratory
DOI: 10.1101/2022.04.05.487135

Generalisation in Neural Networks Does not Require Feature Overlap

Autorzy: Jeff Mitchell; Jeffrey S. Bowers
Opublikowane w: ArXiv, 2021
Wydawca: Cornell University
DOI: 10.48550/arxiv.2107.06872

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

Autorzy: Valerio Biscione; Jeffrey S. Bowers
Opublikowane w: ArXiv, 2021
Wydawca: Cornell University
DOI: 10.48550/arxiv.2110.01476

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

Autorzy: Valerio Biscione, Jeffrey S. Bowers
Opublikowane w: Arxiv, 2022
Wydawca: Arxiv
DOI: 10.48550/arxiv.2203.07302

Deep Problems with Neural Network Models of Human Vision

Autorzy: 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
Opublikowane w: PsyArXiv, 2022
Wydawca: Cornell University
DOI: 10.31234/osf.io/5zf4s

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

Autorzy: Guillermo Puebla; Jeffrey S. Bowers
Opublikowane w: BioRxiv, 2021
Wydawca: Cold Spring Harbor Laboratory
DOI: 10.1101/2021.09.03.458919

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

Autorzy: Gaurav Malhotra; Marin Dujmovic; John Hummel; Jeffrey S Bowers
Opublikowane w: BioRxiv, 2021
Wydawca: Cold Spring Harbor Laboratory
DOI: 10.1101/2021.12.14.472546

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

Autorzy: Milton Montero, Jeffrey S. Bowers, Rui Ponte Costa, Casimir J. Ludwig, Gaurav Malhotra
Opublikowane w: arXiv, 2022
Wydawca: Cornell Univesity
DOI: 10.48550/arxiv.2204.02283

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

Autorzy: Blything, Ryan; Biscione, Valerio; Bowers, Jeffrey
Opublikowane w: arXiv.org e-Print Archive, Numer 1, 2020
Wydawca: arXiv
DOI: 10.48550/arxiv.2012.05950

Learning Translation Invariance in CNNs

Autorzy: Valerio Biscione, Jeffrey Bowers
Opublikowane w: NeurIPS 2020 Workshop SVRHM, 2020
Wydawca: NeurIPS 2020 Workshop SVRHM

The role of disentanglement in generalisation

Autorzy: Milton Llera Montero; Casimir Ludwig; Rui Ponte Costa; Gaurav Malhotra; Jeffrey S Bowers
Opublikowane w: International Conference on Learning Representations, 2021
Wydawca: OpenReview

Can Deep Convolutional Neural Networks Learn Same-Different Relations?

Autorzy: Guillermo Puebla; Jeffrey S. Bowers
Opublikowane w: Proceedings of the Annual Meeting of the Cognitive Science Society, 43(43), Numer 43(43), 2021, ISSN 1069-7977
Wydawca: University of California

Subtractive gating improves generalization in working memory tasks

Autorzy: Milton Llera Montero, Gaurav Malhotra, Jeff Bowers, Rui Ponte Costa
Opublikowane w: 2019 Conference on Cognitive Computational Neuroscience, 2019
Wydawca: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1352-0

Do LSTMs know about Principle C?

Autorzy: Jeff Mitchell, Nina Kazanina, Conor Houghton, Jeff Bowers
Opublikowane w: 2019 Conference on Cognitive Computational Neuroscience, 2019
Wydawca: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1241-0

Extreme Translation Tolerance in Humans and Machines

Autorzy: Ryan Blything, Ivan Vankov, Casimir Ludwig, Jeffrey Bowers
Opublikowane w: 2019 Conference on Cognitive Computational Neuroscience, 2019
Wydawca: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1091-0

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

Autorzy: Marin Dujmović, Gaurav Malhotra, Jeffrey Bowers
Opublikowane w: 2019 Conference on Cognitive Computational Neuroscience, 2019
Wydawca: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1298-0

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

Autorzy: Gaurav Malhotra, Benjamin Evans, Jeffrey Bowers
Opublikowane w: 2019 Conference on Cognitive Computational Neuroscience, 2019
Wydawca: Cognitive Computational Neuroscience
DOI: 10.32470/ccn.2019.1212-0

Translation Tolerance in Vision.

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

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

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

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

Autorzy: Gaurav Malhotra, Jeff Bowers
Opublikowane w: 41st Annual Conference of the Cognitive Science Society 2019, 2019
Wydawca: A.K. Goel, C.M. Seifert, & C. Freksa (Eds.)

Harnessing the Symmetry of Convolutions for Systematic Generalisation

Autorzy: Jeff Mitchell, Jeffrey S. Bowers
Opublikowane w: 2020 International Joint Conference on Neural Networks (IJCNN), 2020, Strona(/y) 1-8, ISBN 978-1-7281-6926-2
Wydawca: IEEE
DOI: 10.1109/ijcnn48605.2020.9207183

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

Autorzy: Christian Tsvetkov, Gaurav Malhotra, Benjamin Evans, Jeff Bowers
Opublikowane w: In Proceedings of the 42nd Annual Conference of the Cognitive Science Society 2020, 2020, Strona(/y) 2358-2364
Wydawca: Cognitive Science Society

Priorless Recurrent Networks Learn Curiously

Autorzy: Jeff Mitchell, Jeffrey Bowers
Opublikowane w: Proceedings of the 28th International Conference on Computational Linguistics, 2020, Strona(/y) 5147-5158, ISBN 978-1-952148-27-9
Wydawca: International Committee on Computational Linguistics
DOI: 10.18653/v1/2020.coling-main.451

Publikacje

Udostępnij tę stronę

Pobierz