Efficient Optimization for Discriminative Latent Class Models
نویسندگان
چکیده
Dimensionality reduction is commonly used in the setting of multi-label supervised classification to control the learning capacity and to provide a meaningful representation of the data. We introduce a simple forward probabilistic model which is a multinomial extension of reduced rank regression, and show that this model provides a probabilistic interpretation of discriminative clustering methods with added benefits in terms of number of hyperparameters and optimization. While the expectation-maximization (EM) algorithm is commonly used to learn these probabilistic models, it usually leads to local maxima because it relies on a non-convex cost function. To avoid this problem, we introduce a local approximation of this cost function, which in turn leads to a quadratic non-convex optimization problem over a product of simplices. In order to maximize quadratic functions, we propose an efficient algorithm based on convex relaxations and lowrank representations of the data, capable of handling large-scale problems. Experiments on text document classification show that the new model outperforms other supervised dimensionality reduction methods, while simulations on unsupervised clustering show that our probabilistic formulation has better properties than existing discriminative clustering methods.
منابع مشابه
Spectral Experts for Estimating Mixtures of Linear Regressions
Discriminative latent-variable models are typically learned using EM or gradient-based optimization, which suffer from local optima. In this paper, we develop a new computationally efficient and provably consistent estimator for a mixture of linear regressions, a simple instance of a discriminative latentvariable model. Our approach relies on a lowrank linear regression to recover a symmetric t...
متن کاملشناسایی نوع و مدل وسیله نقلیه با استفاده از مجموعه بخشهای متمایزکننده
In fine-grained recognition, the main category of object is well known and the goal is to determine the subcategory or fine-grained category. Vehicle make and model recognition (VMMR) is a fine-grained classification problem. It includes several challenges like the large number of classes, substantial inner-class and small inter-class distance. VMMR can be utilized when license plate numbers ca...
متن کاملMedLDA: maximum margin supervised topic models
A supervised topic model can use side information such as ratings or labels associated with documents or images to discover more predictive low dimensional topical representations of the data. However, existing supervised topic models predominantly employ likelihood-driven objective functions for learning and inference, leaving the popular and potentially powerful max-margin principle unexploit...
متن کاملEfficient Latent Structural Perceptron with Hybrid Trees for Semantic Parsing
Discriminative structured prediction models have been widely used in many natural language processing tasks, but it is challenging to apply the method to semantic parsing. In this paper, by introducing hybrid tree as a latent structure variable to close the gap between the input sentences and output representations, we formulate semantic parsing as a structured prediction problem, based on the ...
متن کاملAn application of Measurement error evaluation using latent class analysis
Latent class analysis (LCA) is a method of evaluating non sampling errors, especially measurement error in categorical data. Biemer (2011) introduced four latent class modeling approaches: probability model parameterization, log linear model, modified path model, and graphical model using path diagrams. These models are interchangeable. Latent class probability models express l...
متن کامل