Generalization with Deep Learning
WORLD SCIENTIFIC eBooks2020
Citations Over TimeTop 1% of 2020 papers
Abstract
With a direct analysis of neural networks, this paper presents a mathematically tight generalization theory to partially address an open problem regarding the generalization of deep learning.Unlike previous bound-based theory, our main theory is quantitatively as tight as possible for every dataset individually, while producing qualitative insights competitively.Our results give insight into why and how deep learning can generalize well, despite its large capacity, complexity, possible algorithmic instability, nonrobustness, and sharp minima, answering to an open question in the literature.We also discuss limitations of our results and propose additional open problems.
Related Papers
- → Locating the local minima in lens design with machine learning(2021)3 cited
- → Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applications(2021)3 cited
- → Interpretable Mesomorphic Networks for Tabular Data(2023)1 cited
- A USEFUL BP ALGORITHM TO OVERCOME LOCAL MINIMA(1995)
- Faster escaping from local minima for back-propagation algorithm(2008)