An aggregate and iterative disaggregate algorithm with proven optimality in machine learning

DOI10.1007/S10994-016-5562-ZMaRDI QIDQ1689569zbMATH OpenOpenAlexFDO

Authors Peng Zhang

Publication date 12 January 2018

Published in Machine Learning (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1607.01400

zbMATH Keywords

aggregate and iterative disaggregateaiddata aggregationleast absolute deviation regressionmachine learningoptimizationsemi-supervised support vector machinesupport vector machine

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05) Nonlinear programming (90C30)

Abstract: We propose a clustering-based iterative algorithm to solve certain optimization problems in machine learning, where we start the algorithm by aggregating the original data, solving the problem on aggregated data, and then in subsequent steps gradually disaggregate the aggregated data. We apply the algorithm to common machine learning problems such as the least absolute deviation regression problem, support vector machines, and semi-supervised support vector machines. We derive model-specific data aggregation and disaggregation procedures. We also show optimality, convergence, and the optimality gap of the approximated solution in each iteration. A computational study is provided.

Recommendations

• Aggregation in large-scale optimization
• An algorithm for minimizing clustering functions
• Mathematical programming with iterative aggregation
• On the Convergence Rate of Incremental Aggregated Gradient Algorithms
• scientific article; zbMATH DE number 4023107

Cites work

• Aggregation and Disaggregation in Integer Programming Problems
• Aggregation and Disaggregation Techniques and Methodology in Optimization
• Aggregation in large-scale optimization
• scientific article; zbMATH DE number 3598234 (Why is no real title available?)
• scientific article; zbMATH DE number 2087272 (Why is no real title available?)
• Iterative Aggregation--A New Approach to the Solution of Large-Scale Problems
• LIBLINEAR: a library for large linear classification
• Optimization techniques for semi-supervised support vector machines
• Scikit-learn: machine learning in Python
• Solution of Large-Scale Transportation Problems Through Aggregation
• Solving large-scale linear programs by aggregation
• Solving network design problems via iterative aggregation
• Technical Note—Improved Bounds for Aggregated Linear Programs

Cited in

(7)

• New partial aggregations for multicommodity network flow problems: an application to the fixed-charge network design problem
• Iterative deliberation via metric aggregation
• MILP models for complex system reliability redundancy allocation with mixed components
• Acceleration of machine learning algorithms by Tchebychev iteration technique
• Optimization for L₁-Norm Error Fitting via Data Aggregation
• On Cluster-Aware Supervised Learning: Frameworks, Convergent Algorithms, and Applications
• A learning algorithm for source aggregation

Describes a project that uses

Uses Software

• LIBLINEAR
• R
• quantreg
• LIBSVM
• Scikit

This page was built for publication: An aggregate and iterative disaggregate algorithm with proven optimality in machine learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1689569)