All Publications
2023
“Dynamics in Deep Classifiers Trained with the Square Loss: Normalization, Low Rank, Neural Collapse, and Generalization Bounds”, Research, vol. 6, 2023.
research.0024.pdf (5.19 MB) ,

CBMM Funded
CBMM Memo No.
141
“Feature learning in deep classifiers through Intermediate Neural Collapse”. 2023.
Feature_Learning_memo.pdf (2.16 MB) ,

CBMM Funded
“For interpolating kernel machines, minimizing the norm of the ERM solution maximizes stability”, Analysis and Applications, vol. 21, no. 01, pp. 193 - 215, 2023. ,
CBMM Funded
“Dynamics in Deep Classifiers trained with the Square Loss: normalization, low rank, neural collapse and generalization bounds”, Research, 2023.
research.0024.pdf (4.05 MB) ,

CBMM Funded
CBMM Memo No.
140
“SGD and Weight Decay Provably Induce a Low-Rank Bias in Deep Neural Networks”. 2023.
Low-rank bias.pdf (2.38 MB) ,

CBMM Funded
CBMM Memo No.
139
“Norm-Based Generalization Bounds for Compositionally Sparse Neural Networks”. 2023.
Norm-based bounds for convnets.pdf (1.2 MB) ,

CBMM Funded
2022
“Representation Learning in Sensory Cortex: a theory”, IEEE Access, pp. 1 - 1, 2022.
Representation_Learning_in_Sensory_Cortex_a_theory.pdf (1.17 MB) ,

CBMM Funded
CBMM Memo No.
137
“Understanding the Role of Recurrent Connections in Assembly Calculus”. 2022.
CBMM-Memo-137.pdf (1.49 MB) ,

CBMM Funded
CBMM Memo No.
136
“System identification of neural systems: If we got it right, would we know?”. 2022.
CBMM-Memo-136.pdf (1.75 MB) ,

CBMM Funded
“A computational probe into the behavioral and neural markers of atypical facial emotion processing in autism”, The Journal of Neuroscience, pp. JN-RM-2229-21, 2022. ,
CBMM Related
“Neural Collapse in Deep Homogeneous Classifiers and the role of Weight Decay”, in IEEE International Conference on Acoustics, Speech and Signal Processing, Singapore, 2022. ,
CBMM Funded
“Brain-like functional specialization emerges spontaneously in deep neural networks”, Science Advances, vol. 8, no. 11, 2022. ,
CBMM Funded
CBMM Memo No.
134
“SGD Noise and Implicit Low-Rank Bias in Deep Neural Networks”. 2022.
Implicit Rank Regularization.pdf (1.42 MB) ,

CBMM Funded
“On the Implicit Bias Towards Minimal Depth of Deep Neural Networks”, arXiv, 2022.
2202.09028.pdf (2 MB) ,

CBMM Funded
“Using child‐friendly movie stimuli to study the development of face, place, and object regions from age 3 to 12 years”, Human Brain Mapping, 2022. ,
CBMM Related
CBMM Memo No.
121
“Transformer Module Networks for Systematic Generalization in Visual Question Answering”. 2022.
CBMM-Memo-121.pdf (1.06 MB)
version 2 (3/22/2023) (1.33 MB) ,


CBMM Funded
“When and how convolutional neural networks generalize to out-of-distribution category–viewpoint combinations”, Nature Machine Intelligence, vol. 4, no. 2, pp. 146 - 153, 2022. ,
CBMM Funded
CBMM Memo No.
119
“Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations”. 2022.
CBMM-Memo-119.pdf (31.08 MB) ,

CBMM Funded
CBMM Memo No.
135
“PCA as a defense against some adversaries”. 2022.
CBMM-Memo-135.pdf (2.58 MB) ,

CBMM Funded
“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
“Finding Biological Plausibility for Adversarially Robust Features via Metameric Tasks”, International Conference on Learning Representations (ICLR). 2022. ,
CBMM Funded
2021
“Confidence and central tendency in perceptual judgment”, Attention, Perception, & Psychophysics, vol. 83, no. 7, pp. 3024 - 3034, 2021. ,
CBMM Funded
“Combining Different V1 Brain Model Variants to Improve Robustness to Image Corruptions in CNNs”, in NeurIPS 2021, 2021. ,
CBMM Funded
“Evaluating the Adversarial Robustness of a Foveated Texture Transform Module in a CNN”, NeurIPS 2021. 2021. ,
CBMM Funded
“Computational models of category-selective brain regions enable high-throughput tests of selectivity”, Nature Communications, vol. 12, no. 1, 2021.
s41467-021-25409-6.pdf (6.47 MB) ,

CBMM Funded