Publications

Valerio Biscione, Jeffrey S. Bowers (2021) Learning Online Visual Invariances for Novel Objects via Supervised and Self-Supervised Training (preprint) https://arxiv.org/abs/2110.01476

Valerio Biscione, Jeffrey S. Bowers (2021). Convolutional Neural Networks Are Not Invariant to Translation, but They Can Learn to Be. Journal of Machine Learning Research. 22(229):1−28, 2021.

Guillermo Puebla, Jeffrey Bowers (2021). Can deep convolutional neural networks support relational reasoning in the same-different task?  (preprint) https://doi.org/10.1101/2021.09.03.458919

Jeff Mitchell, Jeffrey S. Bowers (2021). Generalisation in Neural Networks Does not Require Feature Overlap (preprinthttps://arxiv.org/abs/2107.06872

Guillermo Puebla, Jeffrey S. Bowers (2021).  Can Deep Convolutional Neural Networks Learn Same-Different Relations?  (preprint)  https://doi.org/10.1101/2021.04.06.438551

Benjamin D. Evans, Gaurav Malhotra, Jeffrey S. Bowers (2021). Biological convolutions improve DNN robustness to noise and generalisation. (preprint) doi.org/10.1101/2021.02.18.431827

Ryan Blything, Valerio Biscione, Ivan I. Vankov, Casimir J.H. Ludwig, Jeffrey S. Bowers (2021). The human visual system and CNNs can both support robust online translation tolerance following extreme displacements.  Journal of Vision, Vol.21(2):9, 1-16. https://doi.org/10.1167/jov.21.2.9 

Milton Llera Montero, Casimir J. Ludwig, Rui Ponte Costa, Gaurav Malhotra, Jeffrey S. Bowers (2021). The Role of Disentanglement in Generalisation. International Conference on Learning Representations, May 2021. [code]

Ryan Blything, Valerio Biscione, Jeffrey S. Bowers (2020). A case for robust translation tolerance in humans and CNNs. A commentary on Han et al. (preprint) arXiv:2012.05950.

Valerio Biscione and Jeffrey S. Bowers (2020).  Learning Translation Invariance in CNNs. 2nd Workshop on Shared Visual Representations in Human and Machine Intelligence (SVRHM), NeurIPS 2020. https://arxiv.org/abs/2011.11757

Jeff Mitchell & Jeffrey S. Bowers (2020). Priorless Recurrent Networks Learn Curiously. In the Proceedings of the 28th International Conference on Computational Linguistics.  http://dx.doi.org/10.18653/v1/2020.coling-main.451

Marin DujmovićGaurav MalhotraJeffrey S. Bowers, (2020).  What do Adversarial Images Tell us about Human Vision? eLife, https://doi.org/10.7554/eLife.55978 

Christian Tsvetkov, Gaurav Malhotra, Benjamin D. Evans, Jeffrey S. Bowers, (2020) Adding Biological Constraints to Deep Neural Networks Reduces their Capacity to Learn Unstructured Data.  In Proceedings of the 42nd Annual Conference of the Cognitive Science Society 2020, Toronto, Canada.

Ella Gale, Nick D. Martin, Ryan Blything, Anh Tung Nguyen, Jeffrey S. Bowers, (2020) Are there any ‘Object Detectors’ in the Hidden Layers of CNNs Trained to Identify Objects or Scenes? Vision Research, 176, 60-71. https://doi.org/10.1016/j.visres.2020.06.007

Jeff Mitchell and Jeffrey S. Bowers, (2020). Harnessing the Symmetry of Convolutions for Systematic Generalisation, In Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK.  https://doi.org/10.1109/IJCNN48605.2020.9207183 

Gaurav Malhotra, Benjamin Evans and Jeffrey S. Bowers, (2020). Hiding a Plane with a Pixel: Examining Shape-bias in CNNs and the Benefit of Building in Biological Constraints. Vision Research, 174, 57-68. Special Issue on `What do Deep Neural Networks Tell us about Biological Vision’. https://doi.org/10.1016/j.visres.2020.04.013

Ivan Vankov and Jeffrey S. Bowers (2019). Training neural networks to encode symbols enables combinatorial generalization. Philosophical Transactions of the Royal Society. https://doi.org/10.1098/rstb.2019.0309

Marin Dujmović, Gaurav Malhotra and Jeffrey S. Bowers, (2019).  Humans Cannot Decipher Adversarial Images: Revisiting Zhou and Firestone. In Proceedings of the 2019 Conference on Cognitive Computational Neuroscience, Berlin, Germany. https://doi.org/10.32470/CCN.2019.1298-0

Gaurav Malhotra, Benjamin Evans and Jeff S. Bowers, (2019).  Adding Biological Constraints to CNNs Makes Image Classification more Human-Like and Robust. In Proceedings of the 2019 Conference on Cognitive Computational Neuroscience, Berlin, Germany. https://doi.org/10.32470/CCN.2019.1212-0

Ryan Blything, Ivan I. Vankov, Casimir J. Ludwig and Jeffrey S. Bowers, (2019).  Extreme Translation Tolerance in Humans and Machines.  In Proceedings of the 2019 Conference on Cognitive Computational Neuroscience, Berlin, Germany. https://doi.org/10.32470/CCN.2019.1091-0

Milton Llera Montero, Gaurav Malhotra, Jeffrey S. Bowers and Rui Ponte Costa, (2019).  Subtractive Gating Improves Generalization in Working Memory Tasks. In Proceedings of the 2019 Conference on Cognitive Computational Neuroscience, Berlin, Germany. https://doi.org/10.32470/CCN.2019.1352-0

Jeff Mitchell, Nina Kazanina, Conor Houghton and Jeffrey S. Bowers, (2019). Do LSTMs Know about Principle C? In Proceedings of the 2019 Conference on Cognitive Computational Neuroscience, Berlin, Germany.  https://doi.org/10.32470/CCN.2019.1241-0

Ella M. Gale, Ryan Blything, Nick D. Martin, Jeffrey S. Bowers, and Anh Nguyen, (2019).  Selectivity Metrics Provide Misleading Estimates of the Selectivity of Single Units in Neural Networks. In A.K. Goel, C.M. Seifert, & C. Freksa (Eds.), Proceedings of the 41st Annual Conference of the Cognitive Science Society 2019, Montreal, Canada.

Ryan Blything, Ivan I. Vankov, Casimir J. Ludwig and Jeffrey S. Bowers, (2019). Translation Tolerance in Vision. In A.K. Goel, C.M. Seifert, & C. Freksa (Eds.), Proceedings of the 41st Annual Conference of the Cognitive Science Society 2019, Montreal, Canada.

Gaurav Malhotra and Jeffrey S. Bowers, (2019).  The Contrasting Roles of Shape in Human Vision and Convolutional Neural Networks. In A.K. Goel, C.M. Seifert, & C. Freksa (Eds.), Proceedings of the 41st Annual Conference of the Cognitive Science Society 2019, Montreal, Canada.

Jeffrey S. Bowers, Nick D. Martin and Ella M. Gale, (2019). Researchers Keep Rejecting Grandmother Cells after Running the Wrong Experiments: The Issue Is How Familiar Stimuli Are Identified. BioEssays, 2019, 41(8), 1800248.
https://doi.org/10.1002/bies.201800248

Jeffrey S. Bowers, J.S. (2017)Parallel Distributed Processing Theory in the Age of Deep NetworksTrends in Cognitive Science21, 950-961.