Equivalent Wishart Ansatz for kernel renormalization in Bayesian MLPs and CNNs in the proportional regime, with tests showing good agreement on benchmarks.
Predicting kernel regression learning curves from only raw data statistics
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Thermodynamic lower bounds are approximated for exact and SGD linear regression, producing energy-aware scaling laws for optimal training dataset size given a target generalization error.
A mechanics of the learning process is emerging in deep learning theory, characterized by dynamics, coarse statistics, and falsifiable predictions across idealized settings, limits, laws, hyperparameters, and universal behaviors.
citing papers explorer
-
Kernel Renormalization in Bayesian Deep Neural Networks: the Equivalent Wishart Ansatz in the Proportional Regime
Equivalent Wishart Ansatz for kernel renormalization in Bayesian MLPs and CNNs in the proportional regime, with tests showing good agreement on benchmarks.
-
The Thermodynamic Costs of Simple Linear Regression
Thermodynamic lower bounds are approximated for exact and SGD linear regression, producing energy-aware scaling laws for optimal training dataset size given a target generalization error.
-
There Will Be a Scientific Theory of Deep Learning
A mechanics of the learning process is emerging in deep learning theory, characterized by dynamics, coarse statistics, and falsifiable predictions across idealized settings, limits, laws, hyperparameters, and universal behaviors.