Topological properties of basins of attraction of width bounded autoencoders

DOI10.1142/S0219530524500076MaRDI QIDQ6587591zbMATH OpenFDO

Authors H.-P. Beise, Steve Dias Da Cruz

Publication date 14 August 2024

Published in Analysis and Applications (Singapore) (Search for Journal in Brave)

zbMATH Keywords

autoencoderdynamical systembasin of attractionneural network approximationbounded width neural network

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Neural networks for/in biological studies, artificial life and related topics (92B20) Stability of topological dynamical systems (37B25)

Recommendations

• On the benefit of width for neural networks: disappearance of basins
• Complex-valued autoencoders
• scientific article; zbMATH DE number 4092811
• What regularized auto-encoders learn from the data-generating distribution
• Processing simple geometric attributes with autoencoders

Cites work

• A simple neural network generating an interactive memory
• Correlation Matrix Memories
• Deep learning
• Neural networks and physical systems with emergent collective computational abilities
• On Transversality of Bent Hyperplane Arrangements and the Topological Expressiveness of ReLU Neural Networks
• On a model of associative memory with huge storage capacity
• On decision regions of narrow deep neural networks
• Overparameterized neural networks implement associative memory
• Reconciling modern machine-learning practice and the classical bias-variance trade-off
• The capacity of the Hopfield associative memory
• The implicit bias of gradient descent on separable data

This page was built for publication: Topological properties of basins of attraction of width bounded autoencoders

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6587591)