Derives geodesic ridge regularization and Riemannian Gibbs Process prior for feature-learning wide neural networks, generalizing kernel-regime results via function-space axiomatization.
Tuning large neural networks via zero-shot hyperparameter transfer.Advances in Neural Information Processing Systems, 34:17084–17097
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 4roles
background 1polarities
background 1representative citing papers
Neural LoFi models deep learning as layer-wise spectral filtering that selects maximal low-degree correlations, yielding a tractable surrogate for hierarchical representation learning beyond the lazy regime.
Weight decay slows progressive sharpening at the edge of stability, inducing damped oscillations in CNNs and a phase transition to sub-2/η sharpness in MLPs driven by parameter-sharpness gradient alignment, yielding more stable NTK dynamics.
Nora is a matrix optimizer that stabilizes weight norms and angular velocities through row-wise momentum projection onto the orthogonal complement of the weights while approximating structured preconditioning with O(mn) complexity and proven scalability.
citing papers explorer
-
Canonical Regularisation of Wide Feature-Learning Neural Networks
Derives geodesic ridge regularization and Riemannian Gibbs Process prior for feature-learning wide neural networks, generalizing kernel-regime results via function-space axiomatization.
-
Deep Learning as Neural Low-Degree Filtering: A Spectral Theory of Hierarchical Feature Learning
Neural LoFi models deep learning as layer-wise spectral filtering that selects maximal low-degree correlations, yielding a tractable surrogate for hierarchical representation learning beyond the lazy regime.
-
Nora: Normalized Orthogonal Row Alignment for Scalable Matrix Optimizer
Nora is a matrix optimizer that stabilizes weight norms and angular velocities through row-wise momentum projection onto the orthogonal complement of the weights while approximating structured preconditioning with O(mn) complexity and proven scalability.