Binary Linear Classification and Feature Selection via Generalized Approximate Message Passing
Justin Ziniel, Philip Schniter, Per Benjamin Sederberg · IEEE Transactions on Signal Processing · 2015
For the problem of binary linear classification and feature selection, we propose algorithmic approaches to classifier design based on the generalized approximate message passing (GAMP) algorithm, recently proposed in the context of compressive sensing. We are particularly motivated by problems where the number of features greatly exceeds the number of training examples, but where only a few features suffice for accurate classification. We show that sum-product GAMP can be used to (approximately) minimize the classification error rate and max-sum GAMP can be used to minimize a wide variety of regularized loss functions. Furthermore, we describe an expectation-maximization (EM)-based scheme to learn the associated model parameters online, as an alternative to cross-validation, and we show that GAMP's state-evolution framework can be used to accurately predict the misclassification rate. Finally, we present a detailed numerical study to confirm the accuracy, speed, and flexibility afforded by our GAMP-based approaches to binary linear classification and feature selection.
Cited by 22 works. doi:10.1109/tsp.2015.2407311 Open-access PDF
Authors
References
- Maximum Likelihood from Incomplete Data Via the EM Algorithm
- Regularization and Variable Selection Via the Elastic Net
- Wrappers for feature subset selection
- Factor graphs and the sum-product algorithm
- Text Classification from Labeled and Unlabeled Documents using EM
- Pattern Recognition and Machine Learning
- RCV1: A New Benchmark Collection for Text Categorization Research
- Learning Low-Level Vision
- Turbo decoding as an instance of Pearl's "belief propagation" algorithm
- Toward optimal feature selection