Search

In this paper, local learning is proposed to improve the speed and the accuracy of convergence performance of regularity model-based multiobjective estimation of distribution algorithm (RM-MEDA), a typical multi-objective optimization algorithm via estimation of distribution. RM-MEDA employs a model-based method to generate new solutions, however, this method is easy to generate poor solutions when the population has no obvious regularity. To overcome this drawback, our proposed method add a new solution generation strategy, local learning, to the original RM-MEDA. Local learning produces solutions by sampling some solutions from the neighborhood of elitist solutions in the parent population. As it is easy to search some promising solutions in the neighborhood of an elitist solution, local learning can get some useful solutions which help the population attain a fast and accurate convergence. The experimental results on a set of test instances with variable linkages show that the implement of local learning can accelerate convergence speed and add a more accurate convergence to the Pareto optimal. [ABSTRACT FROM AUTHOR]

Abstract: The ultimate goal of distance metric learning is to incorporate abundant discriminative information to keep all data samples in the same class close and those from different classes separated. Local distance metric methods can preserve discriminative information by considering the neighborhood influence. In this paper, we propose a new local discriminative distance metrics (LDDM) algorithm to learn multiple distance metrics from each training sample (a focal sample) and in the vicinity of that focal sample (focal vicinity), to optimize local compactness and local separability. Those locally learned distance metrics are used to build local classifiers which are aligned in a probabilistic framework via ensemble learning. Theoretical analysis proves the convergence rate bound, the generalization bound of the local distance metrics and the final ensemble classifier. We extensively evaluate LDDM using synthetic datasets and large benchmark UCI datasets. [Copyright &y& Elsevier]

Abstract: Topic modeling is a powerful tool for discovering the underlying or hidden structure in text corpora. Typical algorithms for topic modeling include probabilistic latent semantic analysis (PLSA) and latent Dirichlet allocation (LDA). Despite their different inspirations, both approaches are instances of generative model, whereas the discriminative structure of the documents is ignored. In this paper, we propose locally discriminative topic model (LDTM), a novel topic modeling approach which considers both generative and discriminative structures of the data space. Different from PLSA and LDA in which the topic distribution of a document is dependent on all the other documents, LDTM takes a local perspective that the topic distribution of each document is strongly dependent on its neighbors. By modeling the local relationships of documents within each neighborhood via a local linear model, we learn topic distributions that vary smoothly along the geodesics of the data manifold, and can better capture the discriminative structure in the data. The experimental results on text clustering and web page categorization demonstrate the effectiveness of our proposed approach. [Copyright &y& Elsevier]

In this paper, we introduce an approach for the design of electromagnetic devices based on the use of Estimation of Distribution Algorithms (EDAs), coupled with approximation-based local search around the most promising solutions. The main idea is to combine the power of EDAs in the solution of hard optimization problems with the faster convergence provided by the local search using local approximations. The resulting hybrid algorithm is tested on a numerical benchmark problem. [ABSTRACT FROM AUTHOR]

In this paper we introduce an improved implementation of locally weighted projection regression (LWPR), a supervised learning algorithm that is capable of handling high-dimensional input data. As the key features, our code supports multi-threading, is available for multiple platforms, and provides wrappers for several programming languages. [ABSTRACT FROM AUTHOR]

Fuzzy neural network combines the learning capacity of artificial neural networks with the interpretability of the fuzzy systems. A novel structure learning algorithm for fuzzy neural networks (SLNN) is presented in this paper. The neurons of SLNN are created and adapted as online learning proceeds. The learning rule of SLNN is based on Hebb as well as soft competitive learning. The soft competitive learning cannot only let SLNN be able to learn from new data but also prevent it from losing the knowledge that has been learned earlier. To obtain a concise fuzzy rule, a pruning algorithm is adopted in SLNN, which does not disobey the basic design philosophy of fuzzy system. Simulations are performed on the primary benchmarks: circle-in-the-square, two spirals apart, UCI machine learning archive's synthetic control chart time series, and KDDCUP'99 data set. Compared with fuzzy ARTMAP, BP and hierarchical neuro-fuzzy quadtree (HNFQ), the fuzzy neural network achieves higher performance. [ABSTRACT FROM AUTHOR]