All Publications

2023

S. Dae Houlihan, Kleiman-Weiner, M., Hewitt, L. B., Tenenbaum, J. B., and Saxe, R., “Emotion prediction as computation over a generative theory of mind”, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 381, no. 2251, 2023.PDF icon houlihan2023computedappraisals.pdf (2.37 MB)
CBMM Funded
CBMM Memo No.
141
A. Rangamani, Lindegaard, M., Galanti, T., and Poggio, T., “Feature learning in deep classifiers through Intermediate Neural Collapse”. 2023.PDF icon Feature_Learning_memo.pdf (2.16 MB)
CBMM Funded
CBMM Memo No.
140
T. Galanti, Siegel, Z., Gupte, A., and Poggio, T., “SGD and Weight Decay Provably Induce a Low-Rank Bias in Deep Neural Networks”. 2023.PDF icon Low-rank bias.pdf (2.38 MB)
CBMM Funded
CBMM Memo No.
139
T. Galanti, Xu, M., Galanti, L., and Poggio, T., “Norm-Based Generalization Bounds for Compositionally Sparse Neural Networks”. 2023.PDF icon Norm-based bounds for convnets.pdf (1.2 MB)
CBMM Funded

2022

CBMM Memo No.
138
T. A. Poggio, “How Deep Sparse Networks Avoid the Curse of Dimensionality: Efficiently Computable Functions are Compositionally Sparse”. 2022.PDF icon v1.0 (984.15 KB)PDF icon v5.7 adding in context learning etc (1.16 MB)
CBMM Funded
CBMM Memo No.
137
A. Rangamani and Xie, Y., “Understanding the Role of Recurrent Connections in Assembly Calculus”. 2022.PDF icon CBMM-Memo-137.pdf (1.49 MB)
CBMM Funded
CBMM Memo No.
134
T. Galanti and Poggio, T., “SGD Noise and Implicit Low-Rank Bias in Deep Neural Networks”. 2022.PDF icon Implicit Rank Minimization.pdf (1.76 MB)
CBMM Funded
CBMM Memo No.
135
A. Gupte, Banburski, A., and Poggio, T., “PCA as a defense against some adversaries”. 2022.PDF icon CBMM-Memo-135.pdf (2.58 MB)
CBMM Funded
W. Berrios and Deza, A., “Joint rotational invariance and adversarial training of a dual-stream Transformer yields state of the art Brain-Score for Area V4”, in BrainScore Workshop at COSYNE, 2022.
CBMM Funded
A. Harrington and Deza, A., “Finding Biological Plausibility for Adversarially Robust Features via Metameric Tasks”, International Conference on Learning Representations (ICLR). 2022.
CBMM Funded

2021

S. Anzellottti, Houlihan, S. Dae, Liburd, Jr, S., and Saxe, R., “Leveraging facial expressions and contextual information to investigate opaque representations of emotions.”, Emotion, 2021.PDF icon Anzellotti 2021 Emotion.pdf (1.08 MB)
CBMM Funded
CBMM Memo No.
117
M. Xu, Rangamani, A., Banburski, A., Liao, Q., Galanti, T., and Poggio, T., “Dynamics and Neural Collapse in Deep Classifiers trained with the Square Loss”. 2021.PDF icon v1.0 (4.61 MB)PDF icon v1.4corrections to generalization section (5.85 MB)PDF icon v1.7Small edits (22.65 MB)
CBMM Funded
CBMM Memo No.
115
A. Banburski, De La Torre, F., Pant, N., Shastri, I., and Poggio, T., “Distribution of Classification Margins: Are All Data Equal?”. 2021.PDF icon CBMM Memo 115.pdf (9.56 MB)PDF icon arXiv version (23.05 MB)
CBMM Funded
C. Li and Deza, A., “What Matters In Branch Specialization? Using a Toy Task to Make Predictions”, in Shared Visual Representations in Human and Machine Intelligence (SVRHM) Workshop at NeurIPS, 2021.
CBMM Funded
B. Wang, Mayo, D., Deza, A., Barbu, A., and Conwell, C., “On the use of Cortical Magnification and Saccades as Biological Proxies for Data Augmentation”, in Shared Visual Representations in Human and Machine Intelligence (SVRHM) Workshop at NeurIPS, 2021.
CBMM Funded

2020

CBMM Memo No.
113
A. Banburski, Gandhi, A., Alford, S., Dandekar, S., Chin, P., and Poggio, T., “Dreaming with ARC”, Learning Meets Combinatorial Algorithms workshop at NeurIPS 2020. 2020.PDF icon CBMM Memo 113.pdf (1019.64 KB)
CBMM Funded
CBMM Memo No.
109
A. Deza, Liao, Q., Banburski, A., and Poggio, T., “Hierarchically Local Tasks and Deep Convolutional Networks”. 2020.PDF icon CBMM_Memo_109.pdf (2.12 MB)
CBMM Funded

2019

CBMM Memo No.
102
T. Poggio, Kur, G., and Banburski, A., “Double descent in the condition number”. 2019.PDF icon Fixing typos, clarifying error in y, best approach is crossvalidation (837.18 KB)PDF icon Incorporated footnote in text plus other edits (854.05 KB)PDF icon Deleted previous discussion on kernel regression and deep nets: it will appear, extended, in a separate paper (795.28 KB)PDF icon correcting a bad typo (261.24 KB)PDF icon Deleted plot of condition number of kernel matrix: we cannot get a double descent curve  (769.32 KB)
CBMM Funded
CBMM Memo No.
100
T. Poggio, Banburski, A., and Liao, Q., “Theoretical Issues in Deep Networks”. 2019.PDF icon CBMM Memo 100 v1 (1.71 MB)PDF icon CBMM Memo 100 v3 (8/25/2019) (1.31 MB)PDF icon CBMM Memo 100 v4 (11/19/2019) (1008.23 KB)
CBMM Funded
M. Kryven, Scholl, B., and Tenenbaum, J. B., “ Does intuitive inference of physical stability interruptattention?”, Cognitive Sciences Society. 2019.
CBMM Funded
S. J. Gershman, “How to never be wrong”, Psychonomic Bulletin & Review, vol. 26, no. 1, pp. 13 - 28, 2019.
CBMM Funded
M. Araya-Polo, Adler, A., Farris, S., and Jennings, J., “Fast and Accurate Seismic Tomography via Deep Learning”, in Deep Learning: Algorithms and Applications, SPRINGER-VERLAG, 2019.
CBMM Related

2018

T. D. Ullman, Stuhlmüller, A., Goodman, N. D., and Tenenbaum, J. B., “Learning physical parameters from dynamic scenes.”, Cognitive Psychology, vol. 104, pp. 57-82, 2018.PDF icon T-Ullman-etal_CogPsych_LearningPhysicalParametersFromDynamicScenes.pdf (3.15 MB)
CBMM Funded
CBMM Memo No.
095
K. M. Villalobos, Dozier, J., Stih, V., Francl, A., Azevedo, F., Poggio, T., Sasaki, T., and Boix, X., “Can Deep Neural Networks Do Image Segmentation by Understanding Insideness?”. 2018.PDF icon CBMM-Memo-095.pdf (1.96 MB)
CBMM Funded
A. Tacchetti, Voinea, S., and Evangelopoulos, G., “Trading robust representations for sample complexity through self-supervised visual experience”, in Advances in Neural Information Processing Systems 31, Montreal, Canada, 2018, pp. 9640–9650.PDF icon trading-robust-representations-for-sample-complexity-through-self-supervised-visual-experience.pdf (3.32 MB)PDF icon NeurIPS2018_Poster.pdf (6.12 MB)

Pages