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.
Springer, 2008
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Introduces a cross-fitted orthogonal hypergradient estimator derived from the efficient influence function that achieves asymptotic normality and uniform control for bilevel gradient estimation under quadratic losses.
citing papers explorer
-
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.
-
Semiparametric Efficient Bilevel Gradient Estimation
Introduces a cross-fitted orthogonal hypergradient estimator derived from the efficient influence function that achieves asymptotic normality and uniform control for bilevel gradient estimation under quadratic losses.