Principled deep neural network training through linear programming

DOI10.1016/j.disopt.2023.100795zbMath1520.68166arXiv1810.03218OpenAlexW2894812324MaRDI QIDQ6054389

Gonzalo Muñoz, Bienstock, Daniel, Sebastian Pokutta

Publication date: 28 September 2023

Published in: Discrete Optimization (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1810.03218

zbMATH Keywords

linear programmingpolyhedral theorydeep learning

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Combinatorial properties of polytopes and polyhedra (number of faces, shortest paths, etc.) (52B05) Integer programming (90C10) Nonlinear programming (90C30)

Related Items (1)

Optimization of sparsity-constrained neural networks as a mixed integer linear program

Cites Work

• Treewidth. Computations and approximations
• A survey of safety and trustworthiness of deep neural networks: verification, testing, adversarial attack and defence, and interpretability
• Regularisation of neural networks by enforcing Lipschitz continuity
• Lossless compression of deep neural networks
• Neural networks with linear threshold activations: structure and algorithms
• Complexity of training ReLU neural network
• Graph minors. II. Algorithmic aspects of tree-width
• LP Formulations for Polynomial Optimization Problems
• Optimization Methods for Large-Scale Machine Learning
• The Pathwidth and Treewidth of Cographs
• The Computational Complexity of ReLU Network Training Parameterized by Data Dimensionality
• Approximation Algorithms for Training One-Node ReLU Neural Networks
• Understanding Machine Learning
• Strong mixed-integer programming formulations for trained neural networks
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: Principled deep neural network training through linear programming