A Atiya, C Ji - IEEE transactions on neural networks, 1997 - ieeexplore.ieee.org
Generalization is one of the most important problems in neural-network research. It is influenced by several factors in the network design, such as network size, weight decay …
At the heart of machine learning lies the question of generalizability of learned rules over previously unseen data. While over-parameterized models based on neural networks are …
Deep neural networks achieve stellar generalisation even when they have enough parameters to easily fit all their training data. We study this phenomenon by analysing the …
A longstanding goal in deep learning research has been to precisely characterize training and generalization. However, the often complex loss landscapes of neural networks have …
Why does training deep neural networks using stochastic gradient descent (SGD) result in a generalization error that does not worsen with the number of parameters in the network? To …
The classical bias-variance trade-off predicts that bias decreases and variance increase with model complexity, leading to a U-shaped risk curve. Recent work calls this into question for …
This paper investigates the generalization properties of two-layer neural networks in high- dimensions, ie when the number of samples $ n $, features $ d $, and neurons $ h $ tend to …
G Ortiz-Jiménez… - Advances in Neural …, 2021 - proceedings.neurips.cc
For certain infinitely-wide neural networks, the neural tangent kernel (NTK) theory fully characterizes generalization, but for the networks used in practice, the empirical NTK only …
A theoretical understanding of generalization remains an open problem for many machine learning models, including deep networks where overparameterization leads to better …