Publication
Dynamics and Neural Collapse in Deep Classifiers trained with the Square Loss. (2021).
v1.0 (4.61 MB)
v1.4corrections to generalization section (5.85 MB)
v1.7Small edits (22.65 MB)



On the Implicit Bias Towards Minimal Depth of Deep Neural Networks. arXiv (2022). at <https://arxiv.org/abs/2202.09028>
2202.09028.pdf (2 MB)

Feature learning in deep classifiers through Intermediate Neural Collapse. (2023).
Feature_Learning_memo.pdf (2.16 MB)

The Janus effects of SGD vs GD: high noise and low rank. (2023).
Updated with appendix showing empirically that the main results extend to deep nonlinear networks (2.95 MB)
Small updates...typos... (616.82 KB)


Norm-Based Generalization Bounds for Compositionally Sparse Neural Networks. (2023).
Norm-based bounds for convnets.pdf (1.2 MB)

Norm-based Generalization Bounds for Sparse Neural Networks. NeurIPS 2023 (2023). at <https://proceedings.neurips.cc/paper_files/paper/2023/file/8493e190ff1bbe3837eca821190b61ff-Paper-Conference.pdf>
NeurIPS-2023-norm-based-generalization-bounds-for-sparse-neural-networks-Paper-Conference.pdf (577.69 KB)

Formation of Representations in Neural Networks. (2024).
CBMM-Memo-150.pdf (4.03 MB)
