Archetypal landscapes for deep neural networks

DOI10.1073/PNAS.1919995117MaRDI QIDQ5073145zbMATH OpenOpenAlexWikidataFDO

Authors Philipp C. Verpoort, Alpha Albert Lee, D. J. Wales

Publication date 5 May 2022

Published in Proceedings of the National Academy of Sciences (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1073/pnas.1919995117

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07)

Recommendations

• Optimization Landscape of Neural Networks
• A mean field view of the landscape of two-layer neural networks
• Shaping the learning landscape in neural networks around wide flat minima
• Comparing dynamics: deep neural networks versus glassy systems
• Global minima of overparameterized neural networks

Cites work

• scientific article; zbMATH DE number 5621256 (Why is no real title available?)
• A Family of Variable-Metric Methods Derived by Variational Means
• A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
• A mean field view of the landscape of two-layer neural networks
• A new approach to variable metric algorithms
• Cascading classifiers
• Comparing dynamics: deep neural networks versus glassy systems
• Conditioning of Quasi-Newton Methods for Function Minimization
• Entropy-SGD: biasing gradient descent into wide valleys
• Flat Minima
• The Convergence of a Class of Double-rank Minimization Algorithms 1. General Considerations

Cited in

(11)

• Revisiting Landscape Analysis in Deep Neural Networks: Eliminating Decreasing Paths to Infinity
• Community detection-based deep neural network architectures: a fully automated framework based on Likert-scale data
• Shaping the learning landscape in neural networks around wide flat minima
• Comparing dynamics: deep neural networks versus glassy systems
• Spurious valleys in one-hidden-layer neural network optimization landscapes
• Black holes and the loss landscape in machine learning
• Towards interpreting deep neural networks via layer behavior understanding
• Optimization Landscape of Neural Networks
• Landscape and training regimes in deep learning
• Barcodes as summary of loss function topology
• Dissecting a small artificial neural network

This page was built for publication: Archetypal landscapes for deep neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5073145)