Getting away with more network pruning: from sparsity to geometry and linear regions

DOI10.1007/978-3-031-33271-5_14arXiv2301.07966OpenAlexW4377231229MaRDI QIDQ6057261

No author found.

Publication date: 4 October 2023

Published in: Integration of Constraint Programming, Artificial Intelligence, and Operations Research (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2301.07966

zbMATH Keywords

model complexitysolution countingnetwork pruning

Mathematics Subject Classification ID

Combinatorial optimization (90C27) Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20) Operations research and management science (90Bxx)

Cites Work

• Unnamed Item
• Deep neural networks and mixed integer linear optimization
• Multilayer feedforward networks are universal approximators
• Lossless compression of deep neural networks
• Error bounds for approximations with deep ReLU networks
• On mathematical programming with indicator constraints
• Facing up to arrangements: face-count formulas for partitions of space by hyperplanes
• Sensitivity-Informed Provable Pruning of Neural Networks
• JANOS: An Integrated Predictive and Prescriptive Modeling Framework
• Sharp Bounds for the Number of Regions of Maxout Networks and Vertices of Minkowski Sums
• Approximation by superpositions of a sigmoidal function
• Strong mixed-integer programming formulations for trained neural networks

This page was built for publication: Getting away with more network pruning: from sparsity to geometry and linear regions