Search

Traditional supervised learning algorithms do not satisfactorily solve the classification problem on imbalanced data sets, since they tend to assign the majority class, to the detriment of the minority class classification. In this paper, we introduce the Bayesian network-based over-sampling method (BOSME), which is a new over-sampling methodology based on Bayesian networks. Over-sampling methods handle imbalanced data by generating synthetic minority instances, with the benefit that classifiers learned from a more balanced data set have a better ability to predict the minority class. What makes BOSME different is that it relies on a new approach, generating artificial instances of the minority class following the probability distribution of a Bayesian network that is learned from the original minority classes by likelihood maximization. We compare BOSME with the benchmark synthetic minority over-sampling technique (SMOTE) through a series of experiments in the context of indirect cost-sensitive learning, with some state-of-the-art classifiers and various data sets, showing statistical evidence in favor of BOSME, with respect to the expected (misclassification) cost. [ABSTRACT FROM AUTHOR]

In reinforcement learning (RL), the exploration/exploitation (E/E) dilemma is a very crucial issue, which can be described as searching between the exploration of the environment to find more profitable actions, and the exploitation of the best empirical actions for the current state. We focus on the single trajectory RL problem where an agent is interacting with a partially unknown MDP over single trajectories, and try to deal with the E/E in this setting. Given the reward function, we try to find a good E/E strategy to address the MDPs under some MDP distribution. This is achieved by selecting the best strategy in mean over a potential MDP distribution from a large set of candidate strategies, which is done by exploiting single trajectories drawn from plenty of MDPs. In this paper, we mainly make the following contributions: (1) We discuss the strategy-selector algorithm based on formula set and polynomial function. (2) We provide the theoretical and experimental regret analysis of the learned strategy under an given MDP distribution. (3) We compare these methods with the 'state-of-the-art' Bayesian RL method experimentally. [ABSTRACT FROM AUTHOR]

In this paper, we address the problem of direction of arrival (DOA) estimation with coprime array in the context of sparse signal reconstruction to fully exploit the enhanced degrees of freedom (DOF) offered by the difference coarray. The proposed method is based on the framework of sparse Bayesian learning and can jointly refine the unknown DOAs and the sparse signals in a gradual and interweaved manner. Specifically, the proposed approach is constructed by iteratively decreasing a surrogate function majorizing a given objective function, which results in accelerating the speed to converge to the global minimum. Furthermore, for facilitating a noise-free sparse representation, a customized linear transformation is judiciously incorporated in our sparsity-inducing DOA estimator to eliminate the unknown noise variance, and in the mean time, the sample covariance matrix perturbation can be normalized to an identity matrix as a by-product. Extensive simulation experiments under different conditions finally demonstrate the superiority of our suggested algorithm in terms of mean-squared DOA estimation error, DOF and resolution ability over state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Recently, deep unfolding networks have been widely used in direction of arrival (DOA) estimation because of their improved estimation accuracy and reduced computational cost. However, few have considered the existence of a nested array (NA) with off-grid DOA estimation. In this study, we present a deep sparse Bayesian learning (DSBL) network to solve this problem. We first establish the signal model for off-grid DOA with NA. Then, we transform the array output into a real domain for neural networks. Finally, we construct and train the DSBL network to determine the on-grid spatial spectrum and off-grid value, where the loss function is calculated using reconstruction error and the sparsity of network output, and the layers correspond to the steps of the sparse Bayesian learning algorithm. We demonstrate that the DSBL network can achieve better generalization ability without training labels and large-scale training data. The simulation results validate the effectiveness of the DSBL network when compared with those of existing methods. [ABSTRACT FROM AUTHOR]

During the last years, development of open learning environments that support effectively their users has been a challenge for the research community of educational technologies. The open interactive nature of these environments results in users experiencing difficulties in coping with the plethora of available functions, especially during their initial efforts to use the system. In addition, -from the tutors' perspective- the problem solving strategies of the students are often particularly difficult to identify. In this paper, we argue that such problems could be tackled using machine learning techniques such as Bayesian Networks. We show how we can take advantage of log files obtained during field studies to build an adaptive help system providing the most useful support to the student, according to the state of interaction. On the other hand, we attempt to support the tutor, by automating the process of diagnosing students' problem solving strategies using Bayesian Networks. The presented approaches are discussed through examples of two prototypes that have been developed and corresponding evaluation studies. These studies have shown that the proposed approach can effectively support the tasks of students and tutors in such open learning environments. [ABSTRACT FROM AUTHOR]

With the continuous development of new technologies, the scale of training data is also expanding. Machine learning algorithms are gradually beginning to be studied and applied in places where the scale of data is relatively large. Because the current structure of learning algorithms only focus on the identification of dependencies and ignores the direction of dependencies, it causes multiple labeled samples not to identify categories. Multiple labels need to be classified using techniques such as machine learning and then applied to solve the problem. In the environment of more training data, it is very meaningful to explore the structure extension to identify the dependencies between attributes and take into account the direction of dependencies. In this article, Bayesian network structure learning, analysis of the shortcomings of traditional algorithms, and binary evolutionary algorithm are applied to the randomized algorithm to generate the initial population. In the optimization process of the algorithm, it uses a Bayesian network to do a local search and uses a depth-first algorithm to break the loop. Finally, it finds a higher score for the network structure. In the simulation experiment, the classic data sets, ALARM and INSURANCE, are introduced to verify the effectiveness of the algorithm. Compared with NOTEARS and the Expectation-Maximization (EM) algorithm, the weight evaluation index of this article was 4.5% and 7.3% better than other schemes. The clustering effect was improved by 13.5% and 15.2%. The smallest error and the highest accuracy are also better than other schemes. The discussion of Bayesian reasoning in this article has very important theoretical and practical significance. This article further improves the Bayesian network structure and optimizes the performance of the classifier, which plays a very important role in promoting the expansion of the network structure and provides innovative thinking. [ABSTRACT FROM AUTHOR]

One of the main challenges investors have to face is model uncertainty. Typically, the dynamic of the assets is modeled using two parameters: the drift vector and the covariance matrix, which are both uncertain. Since the variance/covariance parameter is assumed to be estimated with a certain level of confidence, we focus on drift uncertainty in this paper. Building on filtering techniques and learning methods, we use a Bayesian learning approach to solve the Markowitz problem and provide a simple and practical procedure to implement optimal strategy. To illustrate the value added of using the optimal Bayesian learning strategy, we compare it with an optimal nonlearning strategy that keeps the drift constant at all times. In order to emphasize the prevalence of the Bayesian learning strategy above the nonlearning one in different situations, we experiment three different investment universes: indices of various asset classes, currencies and smart beta strategies. [ABSTRACT FROM AUTHOR]

Decomposition hybrid algorithms with the recursive framework which recursively decompose the structural task into structural subtasks to reduce computational complexity are employed to learn Bayesian network (BN) structure. Merging rules are commonly adopted as the combination method in the combination step. The direction determination rule of merging rules has problems in using the idea of keeping v-structures unchanged before and after combination to determine directions of edges in the whole structure. It breaks down in one case due to appearances of wrong v-structures, and is hard to operate in practice. Therefore, we adopt a novel approach for direction determination and propose a two-stage combination method. In the first-stage combination method, we determine nodes, links of edges by merging rules and adopt the idea of permutation and combination to determine directions of contradictory edges. In the second-stage combination method, we restrict edges between nodes that do not satisfy the decomposition property and their parent nodes by determining the target domain according to the decomposition property. Simulation experiments on four networks show that the proposed algorithm can obtain BN structure with higher accuracy compared with other algorithms. Finally, the proposed algorithm is applied to the thickening process of gold hydrometallurgy to solve the practical problem. [ABSTRACT FROM AUTHOR]

Abstract: To solve the problem of Bayesian network structure learning with uncertain prior information, a new structure learning method is proposed. In this paper, first the deficiency of current methods that used for gathering prior information is discussed, and a new method of gathering prior information based on belief map is proposed; second minimum description length principle is modified so that it can fuse uncertain prior information; third simulated annealing is modified according to the character of the problem needed to solve and used to search the required structure; last the new structure method is validated by experiments. [Copyright &y& Elsevier]

Bayesian network structure learning from data has been proved to be a NP-hard (Non-deterministic Polynomial-hard) problem. An effective method of improving the accuracy of Bayesian network structure is using experts’ knowledge instead of only using data. Some experts’ knowledge (named here explicit knowledge) can make the causal relationship between nodes in Bayesian Networks (BN) structure clear, while the others (named here vague knowledge) cannot. In the previous algorithms for BN structure learning, only the explicit knowledge was used, but the vague knowledge, which was ignored, is also valuable and often exists in the real world. Therefore we propose a new method of using more comprehensive experts’ knowledge based on hybrid structure learning algorithm, a kind of two-stage algorithm. Two types of experts’ knowledge are defined and incorporated into the hybrid algorithm. We formulate rules to generate better initial network structure and improve the scoring function. Furthermore, we take expert level difference and opinion conflict into account. Experimental results show that our proposed method can improve the structure learning performance. [ABSTRACT FROM AUTHOR]

Oft-decried information overload is a serious problem that negatively impacts the comprehension of information in the digital age. Text summarization is a helpful process that can be used to alleviate this problem. With the aim of seeking a novel method to enhance the performance of multi-document summarization, this study proposes a novel approach to analyze the problem of multi-document summarization based on a mixture model, consisting of a contextual topic model from a Bayesian hierarchical topic modeling family for selecting candidate summary sentences, and a regression model in machine learning for generating the summary. By investigating hierarchical topics and their correlations with respect to the lexical co-occurrences of words, the proposed contextual topic model can determine the relevance of sentences more effectively, recognize latent topics, and arrange them hierarchically. The quantitative evaluation results from a practical application demonstrates that a system implementing this model can significantly improve the performance of summarization and make it comparable to state-of-the-art summarization systems. [ABSTRACT FROM AUTHOR]

Background: The problems of correlation and classification are long-standing in the fields of statistics and machine learning, and techniques have been developed to address these problems. We are now in the era of high-dimensional data, which is data that can concern billions of variables. These data present new challenges. In particular, it is difficult to discover predictive variables, when each variable has little marginal effect. An example concerns Genome-wide Association Studies (GWAS) datasets, which involve millions of single nucleotide polymorphism (SNPs), where some of the SNPs interact epistatically to affect disease status. Towards determining these interacting SNPs, researchers developed techniques that addressed this specific problem. However, the problem is more general, and so these techniques are applicable to other problems concerning interactions. A difficulty with many of these techniques is that they do not distinguish whether a learned interaction is actually an interaction or whether it involves several variables with strong marginal effects. Methodology/Findings: We address this problem using information gain and Bayesian network scoring. First, we identify candidate interactions by determining whether together variables provide more information than they do separately. Then we use Bayesian network scoring to see if a candidate interaction really is a likely model. Our strategy is called MBS-IGain. Using 100 simulated datasets and a real GWAS Alzheimer’s dataset, we investigated the performance of MBS-IGain. Conclusions/Significance: When analyzing the simulated datasets, MBS-IGain substantially out-performed nine previous methods at locating interacting predictors, and at identifying interactions exactly. When analyzing the real Alzheimer’s dataset, we obtained new results and results that substantiated previous findings. We conclude that MBS-IGain is highly effective at finding interactions in high-dimensional datasets. This result is significant because we have increasingly abundant high-dimensional data in many domains, and to learn causes and perform prediction/classification using these data, we often must first identify interactions. [ABSTRACT FROM AUTHOR]

A Bayesian dynamic model based on multitask learning (MTL) is developed for radar automatic target recognition (RATR) using high-resolution range profile (HRRP). The aspect-dependent HRRP sequence is modeled using a truncated stick-breaking hidden Markov model (TSB-HMM) with time-evolving transition probabilities, in which the spatial structure across range cells is described by the hidden Markov structure and the temporal dependence between HRRP samples is described by the time evolution of the transition probabilities. This framework imposes the belief that temporally proximate HRRPs are more likely to be drawn from similar HMMs, while also allowing for possible distant repetition or “innovation”. In addition, as formulated the stick-breaking prior and MTL mechanism are employed to infer the number of hidden states in an HMM and learn the target-dependent states collectively for all targets. The form of the proposed hierarchical model allows efficient variational Bayesian (VB) inference, of interest for large-scale problems. To validate the formulation, example results are presented for an illustrative synthesized dataset and our main application—RATR, for which we consider the measured HRRP data. For the latter, we also make comparisons to the model with the independent state-transition statistics and some other existing statistical models for radar HRRP data. [ABSTRACT FROM PUBLISHER]