Efficient distance metric learning by adaptive sampling and mini-batch stochastic gradient descent (SGD)

DOI10.1007/S10994-014-5456-XzbMATH Open1338.68237arXiv1304.1192OpenAlexW2063135797MaRDI QIDQ493732FDOQ493732

Authors: Qi Qian, Rong Jin, Jinfeng Yi, Shenghuo Zhu, Lijun Zhang 

Publication date: 4 September 2015

Published in: Machine Learning (Search for Journal in Brave)

Abstract: Distance metric learning (DML) is an important task that has found applications in many domains. The high computational cost of DML arises from the large number of variables to be determined and the constraint that a distance metric has to be a positive semi-definite (PSD) matrix. Although stochastic gradient descent (SGD) has been successfully applied to improve the efficiency of DML, it can still be computationally expensive because in order to ensure that the solution is a PSD matrix, it has to, at every iteration, project the updated distance metric onto the PSD cone, an expensive operation. We address this challenge by developing two strategies within SGD, i.e. mini-batch and adaptive sampling, to effectively reduce the number of updates (i.e., projections onto the PSD cone) in SGD. We also develop hybrid approaches that combine the strength of adaptive sampling with that of mini-batch online learning techniques to further improve the computational efficiency of SGD for DML. We prove the theoretical guarantees for both adaptive sampling and mini-batch based approaches for DML. We also conduct an extensive empirical study to verify the effectiveness of the proposed algorithms for DML.

Full work available at URL: https://arxiv.org/abs/1304.1192

Recommendations

• Fast generalization rates for distance metric learning. Improved theoretical analysis for smooth strongly convex distance metric learning
• Adaptive sampling for incremental optimization using stochastic gradient descent
• An efficient method for clustered multi-metric learning
• Survey on distance metric learning and dimensionality reduction in data mining
• Positive semidefinite metric learning using boosting-like algorithms

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Sampling theory, sample surveys (62D05)

Cites Work

• Distance metric learning for large margin nearest neighbor classification
• Title not available (Why is that?)
• Prediction, Learning, and Games
• Large scale online learning of image similarity through ranking
• Title not available (Why is that?)
• Online submodular minimization

Cited In (3)

• Safe triplet screening for distance metric learning
• A fast diagonal distance metric learning approach for large-scale datasets
• Fast generalization rates for distance metric learning. Improved theoretical analysis for smooth strongly convex distance metric learning

Uses Software

• UCI-ml
• LIBSVM
• LMNN

This page was built for publication: Efficient distance metric learning by adaptive sampling and mini-batch stochastic gradient descent (SGD)

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q493732)