Boolean autoencoders and hypercube clustering complexity

DOI10.1007/s10623-012-9719-xzbMath1252.68223MaRDI QIDQ690669

Pierre Baldi

Publication date: 28 November 2012

Published in: Designs, Codes and Cryptography (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10623-012-9719-x

zbMATH Keywords

computational complexity; clustering; learning; Boolean circuits; autoencoders

Mathematics Subject Classification ID

68T05: Learning and adaptive systems in artificial intelligence

94C05: Analytic circuit theory

68P30: Coding and information theory (compaction, compression, models of communication, encoding schemes, etc.) (aspects in computer science)

68Q17: Computational difficulty of problems (lower bounds, completeness, difficulty of approximation, etc.)

Related Items

The capacity of feedforward neural networks

Uses Software

• apcluster
• APCluster
• darch

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Clustering by Passing Messages Between Data Points
• Proof of the squashed cube conjecture
• Cubical graphs and cubical dimensions
• The homeomorphic embedding of \(K_n\) in the \(m\)-cube
• Complex-valued autoencoders
• Reducing the Dimensionality of Data with Neural Networks
• On the Complexity of Some Common Geometric Location Problems
• Deep, Narrow Sigmoid Belief Networks Are Universal Approximators
• The Planar k-Means Problem is NP-Hard
• The complexity of cubical graphs
• On the complexity of some coding problems (Corresp.)
• A Clustering and Data-Reorganizing Algorithm
• On the inherent intractability of certain coding problems (Corresp.)
• The intractability of computing the minimum distance of a code
• Learning representations by back-propagating errors
• A Fast Learning Algorithm for Deep Belief Nets
• $B$-valuations of graphs