Search

Increasing levels of system integration pose difficulties in meeting design specifications for high-performance systems. Oftentimes increased complexity, nonlinearity, and multiple tradeoffs need to be handled simultaneously during the design cycle. Since components in such systems are highly correlated with each other, codesign and co-optimization of the complete system are required. Machine learning (ML) provides opportunities for analyzing such systems with multiple control parameters, where techniques based on Bayesian optimization (BO) can be used to meet or exceed design specifications. In this paper, we propose a new BO-based global optimization algorithm titled Two-Stage BO (TSBO). TSBO can be applied to black box optimization problems where the computational time can be reduced through a reduction in the number of simulations required. Empirical analysis on a set of popular challenge functions with several local extrema and dimensions shows TSBO to have a faster convergence rate as compared with other optimization methods. In this paper, TSBO has been applied for clock skew minimization in 3-D integrated circuits and multiobjective co-optimization for maximizing efficiency in integrated voltage regulators. The results show that TSBO is between $2\times $ - $4\times $ faster as compared with previously published BO algorithms and other non-ML-based techniques. [ABSTRACT FROM AUTHOR]

This paper elaborates on the ranked nodes method (RNM) that is used for constructing conditional probability tables (CPTs) for Bayesian networks consisting of a class of nodes called ranked nodes. Such nodes typically represent continuous quantities that lack well-established interval scales and are hence expressed by ordinal scales. Based on expert elicitation, the CPT of a child node is generated in RNM by aggregating weighted states of parent nodes with a weight expression. RNM is also applied to nodes that are expressed by interval scales. However, the use of the method in this way may be ineffective due to challenges which are not addressed in the existing literature but are demonstrated through an illustrative example in this paper. To overcome the challenges, the paper introduces a novel approach that facilitates the use of RNM. It consists of guidelines concerning the discretization of the interval scales into ordinal ones and the determination of a weight expression and weights based on assessments of the expert about the mode of the child node. The determination is premised on interpretations and feasibility conditions of the weights derived in the paper. The utilization of the approach is demonstrated with the illustrative example throughout the paper. [ABSTRACT FROM PUBLISHER]

Online estimation of the noise floor is essential for the accurate scheduling of wideband code-division multiple-access (WCDMA) enhanced uplink (EUL) channel transmissions. The noise floor is used for computation of the load of the uplink of the cell, which is then used to compute the momentary headroom available for EUL traffic. This paper proves a result on Kalman filter scaling, which significantly extends the dynamic range of a previous noise-floor-estimation scheme. This paper proceeds by proving that an optimal Bayesian minimum estimator can be written in recursive form using geometric--rather than arithmetic--averaging. As compared with previous algorithms, the recursive scheme reduces the memory requirement without compromising performance, at the same time that the tracking properties are enhanced. [ABSTRACT FROM AUTHOR]

Bayesian filtering aims at estimating sequentially a hidden process from an observed one. In particular, sequential Monte Carlo (SMC) techniques propagate in time weighted trajectories which represent the posterior probability density function (pdf) of the hidden process given the available observations. On the other hand, conditional Monte Carlo (CMC) is a variance reduction technique which replaces the estimator of a moment of interest by its conditional expectation given another variable. In this paper, we show that up to some adaptations, one can make use of the time recursive nature of SMC algorithms in order to propose natural temporal CMC estimators of some point estimates of the hidden process, which outperform the associated crude Monte Carlo (MC) estimator whatever the number of samples. We next show that our Bayesian CMC estimators can be computed exactly, or approximated efficiently, in some hidden Markov chain (HMC) models; in some jump Markov state-space systems (JMSS); as well as in multitarget filtering. Finally our algorithms are validated via simulations. [ABSTRACT FROM PUBLISHER]

This paper is devoted to the industrial practices and theoretical approaches for detection and isolation of quality-related multiple faults in large-scale processes. In contrast to the previous schemes, the main innovations are as follows: 1) it is the first time a hierarchical detection and isolation framework for quality-related multiple faults in large-scale processes is developed; 2) a combination method of adaptive kernel canonical variable analysis and Bayesian fusion for real-time and hierarchical detection of varying and unknown quality-related multiple faults is presented; and 3) a robust sparse exponential discriminant analysis algorithm for accurate isolation of multimode quality-related multiple faults is proposed. Finally, the whole framework is applied to a typical large-scale process, i.e., hot strip mill process, where the performance and effectiveness are further demonstrated from real industrial data. [ABSTRACT FROM AUTHOR]

Complementing traditional cellular networks with the option of integrated small cells and WiFi access points can be used to further boost the overall traffic capacity and service level. Small cells along with WiFi access points are projected to carry over 60 percent of all theglobal data traffic by 2015. With the integration of small cells on the radio access network levels, there is a focus on providing operators with more control over small cell selection while reducing the feedback burden. Altogether, these issues motivate the need for innovative distributed and autonomous association policies that operate on each user under the network operator’s control, utilizing only partial information, yet achieving near-optimal solutions for the network. In this paper, we propose a load-aware network selection approach applied to automated dynamic offset in heterogeneous networks (HetNets). In particular, we investigate the properties of a hierarchical (Stackelberg) Bayesian game framework, in which the macro cell dynamically chooses the offset about the state of the channel in order to guide users to perform intelligent network selection decisions between macro cell and small cell networks. We derive analytically the utility related to the channel quality perceived by users to obtain the equilibria, and compare it to the fully centralized (optimal), the full channel state information and the non-cooperative (autonomous) models. Building upon these results, we effectively address the problem of how to intelligently configure a dynamic offset which optimizes network’s global utility while users maximize their individual utilities. One of the technical contributions of the paper lies in obtaining explicit characterizations of the dynamic offset at the equilibrium and the related performances in terms of the price of anarchy. Interestingly, it turns out that the complexity of the algorithm for finding the dynamic offset of the Stackelberg model is \mathcal O(n^4) (where $n$ is the number of users). It is shown that the proposed hierarchical mechanism keeps the price of anarchy almost equal to $1$ even for a low number of users, and remains bounded above by the non-cooperative model. [ABSTRACT FROM PUBLISHER]

Light-field stereo matching problems are commonly modeled by Markov Random Fields (MRFs) for statistical inference of depth maps. Nevertheless, most previous approaches did not adapt to image statistics but instead adopted fixed model parameters. They explored explicit vision cues, such as depth consistency and occlusion, to provide local adaptability and enhance depth quality. However, such additional assumptions could end up confining their applicability, e.g. algorithms designed for dense view sampling are not suitable for sparse one. In this paper, we get back to MRF fundamentals and develop an empirical Bayesian framework—Robust Pseudo Random Field—to explore intrinsic statistical cues for broad applicability. Based on pseudo-likelihoods with hidden soft-decision priors, we apply soft expectation-maximization (EM) for good model fitting and perform hard EM for robust depth estimation. We introduce novel pixel difference models to enable such adaptability and robustness simultaneously. Accordingly, we devise a stereo matching algorithm to employ this framework on dense, sparse, and even denoised light fields. It can be applied to both true-color and grey-scale pixels. Experimental results show that it estimates scene-dependent parameters robustly and converges quickly. In terms of depth accuracy and computation speed, it also outperforms state-of-the-art algorithms constantly. [ABSTRACT FROM AUTHOR]

The article discusses various papers published within issue, including oe by Forestiero and Mastroianni on swarm algorithm for self-structured P2P information system and another by Pinto et al. on using local discovery ant algorithm for Bayesian network structure learning.

Multi-label classification is an important research topic in machine learning, for which exploiting label dependencies is an effective modeling principle. Recently, probabilistic models have shown great potential in discovering dependencies among labels. In this paper, motivated by the recent success of multi-view learning to improve the generalization performance, we propose a novel multi-view probabilistic model named latent conditional Bernoulli mixture (LCBM) for multi-label classification. LCBM is a generative model taking features from different views as inputs, and conditional on the latent subspace shared by the views a Bernoulli mixture model is adopted to build label dependencies. Inside each component of the mixture, the labels have a weak correlation which facilitates computational convenience. The mean field variational inference framework is used to carry out approximate posterior inference in the probabilistic model, where we propose a Gaussian mixture variational autoencoder (GMVAE) for effective posterior approximation. We further develop a scalable stochastic training algorithm for efficiently optimizing the model parameters and variational parameters, and derive an efficient prediction procedure based on greedy search. Experimental results on multiple benchmark datasets show that our approach outperforms other state-of-the-art methods under various metrics. [ABSTRACT FROM AUTHOR]

The Gaussian mixture probability hypothesis density (GM-PHD) filter has been widely adopted to track multiple targets, because it can effectively handle target birth/death without the track-to-measurement data association process. However, the GM-PHD filter is known to have serious problems related to birth intensity generation and target tractability. In addition, weight underestimation/overestimation may occur if there are missing detections or measurement clutters. Since these problems may lead to severe estimation errors, many researchers have tried to find solutions. However, none of the researchers have been successful at solving these problems simultaneously. In this paper, we propose a robust multitarget tracking scheme based on the GM-PHD filter to improve estimation accuracy, even if there are many false detections. The proposed scheme includes the processing step of evaluating multiple states/measurements, which is designed to overcome the weight underestimation/overestimation problems. Furthermore, it includes generating the birth intensity for the next iteration using measurements not associated with any tracked states. We also show that the proposed method can be extended to nonlinear Gaussian models. The simulation results demonstrate that the proposed scheme can provide relatively accurate multitarget estimates compared with the previous approaches when the measurements include many false positives/negatives. [ABSTRACT FROM PUBLISHER]

It is known that in addition to spectrum sparsity, spatial sparsity can also be used to further enhance spectral utilization in cognitive radio systems. To achieve that, secondary users (SUs) must know the locations and signal strength distributions (SSDs) of primary users' base stations (PUBSs). Recently, a group sparse total least squares method was developed to cooperatively sense the PUBSs' signal strength and estimate their locations. It approximates PUBSs' power decay with a path loss model (PLM), assumes PUBSs' locations on some grid points, and then accomplishes the estimation tasks. However, the parameters of the PLM have to be known in advance, and the accuracy of the location estimation is bounded by the resolution of the grid points, which limit its practical applications. In this paper, we propose a sparse Bayesian learning method to solve the problems. We use a Laplacian function to model the power decay of a PUBS and then derive learning rules to estimate corresponding parameters. The distinct features of the proposed method are that most parameters are adaptively estimated, and little prior information is needed. To further enhance the performance, we incorporate source number detection methods in the proposed algorithm such that the number of the PUBSs can be precisely detected, facilitating the estimation of PUBSs' locations and SSDs. Moreover, the proposed algorithm is modified into a recursive mode to adapt to SUs' mobility and time-variant observations. Simulations show that the proposed algorithm has good performance, even when the spatial measurement rate is low. [ABSTRACT FROM PUBLISHER]

A time or spatial series of drop counts is but one realization of a multiple stochastic process. In this paper, a method is presented that extracts more of the information contained in the time series of 1-min Joss-Waldvogel disdrometer counts in rain than a simple analysis of the magnitudes of the counts would provide. This is done by greatly increasing the size of a data set using a Bayesian analysis of drop count measurements in 17 size bins. Using the empirical copula statistical technique of probability density function transformations, a 1391-min time series of drop counts was expanded to the equivalent of 40 000 min. This dramatic increase in sample size permits a deeper characterization of the rain. Using this single disdrometer, it also allows one to translate these counts into a 200 × 200 grid filled at each point with drop size distributions of mean drop concentrations consistent with the observed statistical properties of the rain. Such a field can be used for remote sensing studies of the effect of partial beam filling and for algorithm development. Moreover, since there is nothing unique to this set of drop counts, this approach can be applied to any other set of count data, including snow and clouds. [ABSTRACT FROM PUBLISHER]

Parameters tuning of the model and the controller is an essential problem for autonomous vehicles. Traditionally, parameter tuning work is accomplished by manual operation or grid search, which is a tedious and time-consuming work. Recently, thanks to the development of machine learning community, several automatic controller parameter tuning approaches emerged, which usually model the performance function as a Gaussian process, and complete the automatic tuning procedure via Bayesian optimization. However, the existing approaches rarely consider the time-varying feature of the system performance, which is practical in many scenarios, induced by the unmodeled interactions between the system and the environment. In this paper, we take both of the dynamic model uncertainty and the controller parameters uncertainty into account, and tune them to find a global optimal choice for minimizing the time-varying control costs, which is modeled by the time-varying Gaussian process. We provide a novel algorithm, named as time-varying controller optimization. We validate our approach on synthetic simulation and real experiment, respectively. Taking the cumulative regret as the performance metric, we find that our approach has a better performance compared with the stationary algorithm. [ABSTRACT FROM AUTHOR]

This paper proposes an online multiple object tracker that can operate under unknown detection profile and clutter rate. In a majority of multiple object tracking applications, model parameters for background processes such as clutter and detection are unknown and vary with time; hence, the ability of the algorithm to adaptively learn these parameters is essential in practice. In this paper, we detail how the generalized labeled multibernoulli filter, a tractable and provably Bayes optimal multiobject tracker, can be tailored to learn clutter and detection parameters on-the-fly while tracking. Provided that these background model parameters do not fluctuate rapidly compared to the data rate, the proposed algorithm can adapt to the unknown background yielding better tracking performance. [ABSTRACT FROM AUTHOR]

This paper describes the adaptation of a Bayesian sea ice detection algorithm for the scatterometer on-board the European Remote Sensing (ERS) satellites (ERS-1 and ERS-2). The algorithm is based on statistics of distances to ocean wind and sea ice geophysical model functions (GMFs) and its performance is validated against coincident active and passive microwave data. We furthermore propose a new model for sea ice backscatter at the C-band in vertical polarization based on the sea ice GMFs derived from ERS and advanced scatterometer data. The model characterizes the dependence of sea ice backscatter on the incidence angle and the sea ice type, allowing a more precise incidence angle correction than afforded by the usual linear transformation. The resulting agreement between the ERS, QuikSCAT, and special sensor microwave imager sea ice extents during the year 2000 is high during the fall and winter seasons, with an estimated ice edge accuracy of about 20 km, but shows persistent biases between scatterometer and radiometer extents during the melting period, with scatterometers being more sensitive to summer (lower concentration and rotten) sea ice types. [ABSTRACT FROM AUTHOR]

Video segmentation is a task of partitioning pixels that exhibit homogeneous appearance and motion into coherent spatial–temporal groups, which is still challenging for aerial applications. In this paper, a principled combination of superpixel labeling priors and Markov random field (S-MRF) is proposed for aerial video segmentation. The proposed approach has several contributions: 1) we develop a metadata-based global projection model with coordinate transformation to estimate motion information between frames; 2) the superpixel labeling priors from previous frames are incorporated into the segmentation of the current frame, leading to a highly efficient probabilistic label propagation algorithm; and 3) we perform an MRF optimization on the initial segments with propagated labeling priors to improve the temporal coherency. In addition, a new video dataset is collected and will be made publicly available to evaluate the performance of aerial video segmentation algorithms. The experimental results show that the proposed approach outperforms the state-of-the-art video segmentation methods. [ABSTRACT FROM AUTHOR]

Direction of arrival (DOA) estimation is a classical problem in signal processing with many practical applications. Its research has recently been advanced owing to the development of methods based on sparse signal reconstruction. While these methods have shown advantages over conventional ones, there are still difficulties in practical situations where true DOAs are not on the discretized sampling grid. To deal with such an off-grid DOA estimation problem, this paper studies an off-grid model that takes into account effects of the off-grid DOAs and has a smaller modeling error. An iterative algorithm is developed based on the off-grid model from a Bayesian perspective while joint sparsity among different snapshots is exploited by assuming a Laplace prior for signals at all snapshots. The new approach applies to both single snapshot and multi-snapshot cases. Numerical simulations show that the proposed algorithm has improved accuracy in terms of mean squared estimation error. The algorithm can maintain high estimation accuracy even under a very coarse sampling grid. [ABSTRACT FROM PUBLISHER]

A novel framework using a Bayesian approach for content-based phishing web page detection is presented. Our model takes into account textual and visual contents to measure the similarity between the protected web page and suspicious web pages. A text classifier, an image classifier, and an algorithm fusing the results from classifiers are introduced. An outstanding feature of this paper is the exploration of a Bayesian model to estimate the matching threshold. This is required in the classifier for determining the class of the web page and identifying whether the web page is phishing or not. In the text classifier, the naive Bayes rule is used to calculate the probability that a web page is phishing. In the image classifier, the earth mover's distance is employed to measure the visual similarity, and our Bayesian model is designed to determine the threshold. In the data fusion algorithm, the Bayes theory is used to synthesize the classification results from textual and visual content. The effectiveness of our proposed approach was examined in a large-scale dataset collected from real phishing cases. Experimental results demonstrated that the text classifier and the image classifier we designed deliver promising results, the fusion algorithm outperforms either of the individual classifiers, and our model can be adapted to different phishing cases. [ABSTRACT FROM AUTHOR]

We present in this paper an integrated solution to rapidly recognizing dynamic objects in surveillance videos by exploring various contextual information. This solution consists of three components. The first one is a multi-view object representation. It contains a set of deformable object templates, each of which comprises an ensemble of active features for an object category in a specific view/pose. The template can be efficiently learned via a small set of roughly aligned positive samples without negative samples. The second component is a unified spatio-temporal context model, which integrates two types of contextual information in a Bayesian way. One is the spatial context, including main surface property (constraints on object type and density) and camera geometric parameters (constraints on object size at a specific location). The other is the temporal context, containing the pixel-level and instance-level consistency models, used to generate the foreground probability map and local object trajectory prediction. We also combine the above spatial and temporal contextual information to estimate the object pose in scene and use it as a strong prior for inference. The third component is a robust sampling-based inference procedure. Taking the spatio-temporal contextual knowledge as the prior model and deformable template matching as the likelihood model, we formulate the problem of object category recognition as a maximum-a-posteriori problem. The probabilistic inference can be achieved by a simple Markov chain Mento Carlo sampler, owing to the informative spatio-temporal context model which is able to greatly reduce the computation complexity and the category ambiguities. The system performance and benefit gain from the spatio-temporal contextual information are quantitatively evaluated on several challenging datasets and the comparison results clearly demonstrate that our proposed algorithm outperforms other state-of-the-art algorithms. [ABSTRACT FROM PUBLISHER]

Detection and delineation of P- and T-waves are important issues in the analysis and interpretation of electrocardiogram (ECG) signals. This paper addresses this problem by using Bayesian inference to represent a priori relationships among ECG wave components. Based on the recently introduced partially collapsed Gibbs sampler principle, the wave delineation and estimation are conducted simultaneously by using a Bayesian algorithm combined with a Markov chain Monte Carlo method. This method exploits the strong local dependency of ECG signals. The proposed strategy is evaluated on the annotated QT database and compared to other classical algorithms. An important feature of this paper is that it allows not only for the detection of P- and T-wave peaks and boundaries, but also for the accurate estimation of waveforms for each analysis window. This can be useful for some ECG analysis that require wave morphology information. [ABSTRACT FROM AUTHOR]

In the upcoming Internet-of-Things (IoT) era, the communication is often featured by massive connection, sporadic transmission, and small-sized data packets, which poses new requirements on the delay expectation and resource allocation efficiency of the Random Access (RA) mechanisms of the IoT communication stack. A grant-free non-orthogonal random access (NORA) system is considered in this paper, which could simultaneously reduce the access delay and support more Machine Type Communication (MTC) devices with limited resources. In order to address the joint user activity detection (UAD) and channel estimation (CE) problem in the grant-free NORA system, we propose a deep neural network-aided message passing-based block sparse Bayesian learning (DNN-MP-BSBL) algorithm. In the DNN-MP-BSBL algorithm, the iterative message passing process is transferred from a factor graph to a deep neural network (DNN). Weights are imposed on the messages in the DNN and trained to minimize the estimation error. It is shown that the trained weights could alleviate the convergence problem of the MP-BSBL algorithm, especially on crowded RA scenarios. Simulation results show that the proposed DNN-MP-BSBL algorithm could improve the UAD and CE accuracy with a smaller number of iterations, indicating its advantages for low-latency grant-free NORA systems. [ABSTRACT FROM AUTHOR]

An evolutionary algorithm-based approach to identify effective courses of action (COAs) in dynamic uncertain situations is presented. The uncertain situation is modeled using timed influence nets, an instance of dynamic Bayesian networks. The approach makes significant enhancements to the current trial-and-error-based manual technique, which is not only labor intensive but also not capable of modeling constraints among actionable events. The proposed approach is an attempt to overcome these limitations. It automates the process of COA identification. It also allows a system analyst to capture certain types of constraints among actionable events. Because of its parallel search nature, the approach produces multiple COAs that have a similar fitness value. This feature not only gives more flexibility to a decision maker during mission planning, but it can also be used to generalize the COAs if there exists a pattern among them. This paper also discusses a heuristic that further enhances the performance of the approach. [ABSTRACT FROM AUTHOR]

This paper addresses the capability of synthetic aperture radar and optical images in combination with theoretical models to detect the vegetation water content (VWC) at field level. In this paper, a retrieval algorithm for the estimation of VWC from AirSAR acquired on vegetated fields during the SMEX'02 experiment is addressed. The aforementioned campaign has been chosen because, along with sensor observations, extensive ground truth measurements were acquired. The retrieval procedure, which is based on a Bayesian approach, has been initially developed for soil moisture extraction, It consists of two modules: one is pertinent to bare soils and the other one has been modified for vegetated fields. The last one uses the synergy with optical images to correct for the contribution of VWC. The VWC, a variable in the inversion procedure, as well as soil moisture can be estimated. The results indicate a good correlation with both ground measurements and VWC calculated from Landsat images through the use of normalized difference water index (NDWI). Furthermore, in the inversion procedure, the introduction of the dependence on roughness improves the estimates. This indicates that, even for dense vegetation, the contribution from bare soil greatly influences the radar signal. Three main levels of VWC are discriminated in the inversion procedure: values below 1 kg/m², values between 1 and 3 kg/m², and values greater than 3 kg/m². [ABSTRACT FROM AUTHOR]

This paper considers a Bayesian model-averaging (MA) approach to learn an unsupervised naive Bayes classification model. By using the expectation model-averaging (EMA) algorithm, which is proposed in this paper, a unique naive Bayes model that approximates an MA over selective naive Bayes structures is obtained. This algorithm allows to obtain the parameters for the approximate MA clustering model in the same time complexity needed to learn the maximum-likelihood model with the expectation-maximization algorithm. On the other hand, the proposed method can also be regarded as an approach to an unsupervised feature subset selection due to the fact that the model obtained by the EMA algorithm incorporates information on how dependent every predictive variable is on the cluster variable. [ABSTRACT FROM AUTHOR]

Grid computing is a new emerging technology aiming at large-scale resource sharing, and global-area collaboration. It is a next step in an evolution of parallel and distributed computing. Due to the large scale and complexity of the grid system, its performance and reliability are difficult to model, analyse, and evaluate. This paper presents a model that relaxes some assumptions unsuitable for grid computing systems that have been made in the existed works studying the distributed systems. The paper proposes a virtual tree model of the grid service. This model simplifies the physical structure of a grid service, allows service performance (execution time) to be estimated, and takes into account the common cause failures in communication channels. Based on the model, an algorithm for evaluating the grid service performance distribution and the service reliability indices is suggested. The algorithm is based on graph theory, and Bayesian analysis. Illustrative examples are presented in which the results of the suggested algorithm are compared with simulation results. [ABSTRACT FROM AUTHOR]

When reconstructing SPECT data using the maximum a posteriori algorithm with entropy-based prior (MAP-ENT), the effectiveness of the algorithm depends strongly upon the choice of regularization parameter. The problem of choosing an optimal parameter remains still an open question in SPECT. Convergence properties Of the MAP-ENT algorithm depending on a regularization parameter are investigated in this paper. The main goal was to study the effect of iteratively adjusting the regularization parameter on stability of the reconstruction process. Numerical tests have shown that there is a tradeoff between the resolution and noise level of the image that changes with iterations. It entails the choice of an appropriate regularization parameter at each iteration step. The iterative regularization was performed in the following way: the "best" regularization parameter was selected at each iteration step by using the adaptive chi-square criterion. A comparison with the maximum-likelihood-based OS-EM algorithm has shown that both (OS-EM and MAP-ENT) algorithms rather confidently lead the solution to a `feasible' image in a few iterations. Then, as the iteration number increases, the OS-EM solution passes the feasibility area and produces images that begin to deteriorate. The MAP-ENT algorithm retains the solution within the interval due to the dynamic choice of an appropriate regularization parameter. [ABSTRACT FROM AUTHOR]

In this paper, we present two efficient strategies for likelihood computation and decoding in a continuous speech recognizer using an underlying nonlinear state-space dynamic model for the hidden speech dynamics. The state-space model has been specially constructed so as to be suitable for the conversational or casual style of speech where phonetic reduction abounds. Two specific decoding algorithms, based on optimal state-sequence estimation for the nonlinear state-space model, are derived, implemented, and evaluated. They successfully overcome the exponential growth in the original search paths by using the path-merging approaches derived from Bayes' rule. We have tested and compared the two algorithms using the speech data from the Switchboard corpus, confirming their effectiveness. Conversational speech recognition experhyments using the Switchboard corpus further demonstrated that the use of the new decoding strategies is capable of reducing the recognizer's word error rate compared with two baseline recognizers, including the HMM system and the nonlinear state-space model using the HMM-produced phonetic boundaries, under identical test conditions. [ABSTRACT FROM AUTHOR]

This paper presents a new Bayesian model and algorithm used for depth and reflectivity profiling using full waveforms from the time-correlated single-photon counting measurement in the limit of very low photon counts. The proposed model represents each Lidar waveform as a combination of a known impulse response, weighted by the target reflectivity, and an unknown constant background, corrupted by Poisson noise. Prior knowledge about the problem is embedded through prior distributions that account for the different parameter constraints and their spatial correlation among the image pixels. In particular, a gamma Markov random field (MRF) is used to model the joint distribution of the target reflectivity, and a second MRF is used to model the distribution of the target depth, which are both expected to exhibit significant spatial correlations. An adaptive Markov chain Monte Carlo algorithm is then proposed to perform Bayesian inference. This algorithm is equipped with a stochastic optimization adaptation mechanism that automatically adjusts the parameters of the MRFs by maximum marginal likelihood estimation. Finally, the benefits of the proposed methodology are demonstrated through a series of experiments using real data. [ABSTRACT FROM AUTHOR]

Subspace-based high-resolution algorithms for direction-of-arrival (DOA) estimation have been developed for large-scale adaptive antenna arrays. However, its prerequisite step, namely, source enumeration, has not yet been addressed. In this paper, a new approach is devised in the framework of the Bayesian information criterion (BIC) to provide reliable detection of the signal source number for the general asymptotic regime, where m,n\rightarrow\infty and m/n\rightarrow c\in(\mbox{0},\infty), with m$ and $n$ being the numbers of antennas and snapshots, respectively. In particular, the a posteriori probability is determined by correctly calculating the LLFs and PFs for the general asymptotic case. By means of the maximum a posteriori probability, we are capable of effectively finding the signal number. An accurate closed-form expression for the probability of missed detection is also derived for the proposed BIC variant. In addition, the probability of false alarm for the BIC detector is proved to converge to zero as $m,n\!\rightarrow\!\infty$ and $m/n\!\rightarrow\! c$. Simulation results are included to demonstrate the superiority of the proposed detection approach over state-of-the-art schemes and corroborate our theoretical calculations. [ABSTRACT FROM AUTHOR]

This paper considers the problem of multiple human target tracking in a sequence of video data. A solution is proposed which is able to deal with the challenges of a varying number of targets, interactions, and when every target gives rise to multiple measurements. The developed novel algorithm comprises variational Bayesian clustering combined with a social force model, integrated within a particle filter with an enhanced prediction step. It performs measurement-to-target association by automatically detecting the measurement relevance. The performance of the developed algorithm is evaluated over several sequences from publicly available data sets: AV16.3, CAVIAR, and PETS2006, which demonstrates that the proposed algorithm successfully initializes and tracks a variable number of targets in the presence of complex occlusions. A comparison with state-of-the-art techniques due to Khan , Laet , and Czyz shows improved tracking performance. [ABSTRACT FROM PUBLISHER]

In comparison with H.264/Advanced Video Coding, the newest video coding standard, High Efficiency Video Coding (HEVC), improves video coding rate-distortion (RD) performance, but at the price of significant increase in its encoding complexity, especially, in intra-mode decision due to the adoption of more complex block partitions and more candidate intra-prediction modes (IPMs). To reduce the mode decision complexity in HEVC intra-frame coding, while maintaining its RD performance, in this paper, we first formulate the mode decision problem in intra-frame coding as a Bayesian decision problem based on the newly proposed transparent composite model (TCM) for discrete cosine transform coefficients, and then present an outlier-based fast intra-mode decision (OIMD) algorithm. The proposed OIMD algorithm reduces the complexity using outliers identified by TCM to make a fast coding unit split/nonsplit decision and reduce the number of IPMs to be compared. To further take advantage of the outlier information furnished by TCM, we also refine entropy coding in HEVC by encoding the outlier information first, and then the actual mode decision conditionally given the outlier information. The proposed OIMD algorithm can work with and without the proposed entropy coding refinement. Experiments show that for the all-intra-main test configuration of HEVC: 1) when applied alone, the proposed OIMD algorithm reduces, on average, the encoding time (ET) by 50% with 0.7% Bjontegaard distortion (BD)-rate increase and 2) when applied in conjunction with the proposed entropy coding refinement, it reduces, on average, both the ET by 50% and BD-rate by 0.15%. [ABSTRACT FROM PUBLISHER]

This paper develops a novel algorithm for classifying and rejecting non-line-of-sight ultrasonic signals for use in indoor positioning and measurement systems. The algorithm consists of two parts, an initial probability estimate based upon signal amplitude followed by an iteratively reweighed least squares regression that utilizes the probabilities in the weight update step. This method allows for reflections to be identified and rejected while simultaneously calculating the position of a receiving node. The algorithm has been tested on a data collected during experiments designed to generate many challenging specular reflections. The experiments were conducted with transducers in a variety of different positions, and amplitude-modulated ultrasonic signals were used with four different envelopes. The algorithm was capable of correctly classifying over 96% of narrowband signals with severe multipath effect at low computational cost. [ABSTRACT FROM AUTHOR]

In this paper, a computationally efficient algorithm for Bayesian joint change point detection (CPD) in multiple time series is presented. The data generation model includes a number of change configurations (CC), each affecting a unique subset of the time series, which introduces correlation between the positions of change points (CPs) in the monitored time series. The inference objective is to identify joint changes and the associated CC. The algorithm consists of two stages: First, a univariate CPD algorithm is applied separately to each of the involved time series. The outcomes of this step are maximum a posteriori (MAP) detected CPs and posterior distributions of CPs conditioned on the MAP CPs. These outcomes are used in combination to approximate the posterior for the CCs. In the second algorithm stage, dynamic programming is used to find the maxima of this approximate CC posterior. The algorithm is applied to synthetic data, and it is shown to be both significantly faster and more accurate compared to a previously proposed algorithm designed to solve similar problems. Also, the initial algorithm is extended with steps from the maximization–maximization algorithm, which allows the hyperparameters of the data generation model to be estimated jointly with the CCs, and we show that these estimates coincide with estimates obtained from a Markov chain Monte Carlo algorithm. [ABSTRACT FROM AUTHOR]

In this paper, we present an approach to analyze a process workflow for the fabrication of semiconductor integrated circuits in order to identify the source(s) of yield loss. Specifically, we address computational challenges using Bayesian analysis to isolate key process steps affecting the global yield loss of a given workflow. Bayesian inference algorithms incur a heavy computational penalty when applied to large graphs with repetitions, un-modeled interactions, and reentrants, all of which are commonly found in semiconductor manufacturing workflows. We are able to improve the performance of the diagnostic inference process by “folding” and “clustering” nodes, where possible, to simplify the analysis. We also include the effect of un-modeled interactions called latent effects by using a memory efficient auxiliary relaxation. Finally, we present a method to strategically reverse the simplifications, when necessary, so that the accuracy of the inference result is not compromised. The analysis process is shown to be effective at identifying the root cause of global yield discrepancies in a semiconductor manufacturing workflow. We also show that auxiliary relaxation improves the convergence properties of the inference algorithm and reduces the net uncertainty and noise in the final result. [ABSTRACT FROM AUTHOR]

Due to its ability of realizing localization without the need of equipping the target with a wireless device, the device-free wireless localization technique has become a crucial technique for many security and military applications. However, there still lacks an efficient scheme which could achieve robust location estimation performance with low computational cost. To solve this problem, we propose a lightweight robust Bayesian grid approach (BGA) in this paper. The BGA utilizes not only the observation information of the shadowed links, but also the prior information involved in the previous estimations and the constraint information involved in the non-shadowed links, which ensure its robust performance. Meanwhile, the BGA can be carried out with a series of lightweight grid multiplication and addition operations, which eliminates the complex matrix inversion computation involved in the traditional algorithm. The experimental results demonstrate that BGA could achieve a mean tracking error of 0.155 m with a running time of only 1.5 ms. [ABSTRACT FROM PUBLISHER]

Recently, compressive sensing (CS) has been successfully used in inverse synthetic aperture radar (ISAR) imaging. Since the exact sparse reconstruction, i.e., l0-norm constraint, is NP hard, l1-norm relaxation is widely used at the cost of performance degradation in the sparseness of the solution. The performance of existing CS-based ISAR imaging algorithms is sensitive to the regularized factor, which should be adjusted manually. This makes the existing algorithms inconvenient to be used in practice. It is well known that sparse Bayesian learning (SBL) acts as an effective tool in regression and classification, which is closely related to the CS. Furthermore, all the necessary parameters can be estimated using an efficient evidence maximization procedure in SBL, which retains a preferable property of the l0-norm diversity measure and can give more sparse solution. Motivated by that, a fully automated ISAR imaging algorithm based on SBL is proposed in this paper. Experimental results based on simulated and measured data show that the proposed algorithm keeps a better balance between the computation load and the sparsity of the reconstruction signal than the existing algorithms. [ABSTRACT FROM PUBLISHER]

This paper proposes a new multiobjective estimation of distribution algorithm (EDA) based on joint probabilistic modeling of objectives and variables. This EDA uses the multidimensional Bayesian network as its probabilistic model. In this way, it can capture the dependencies between objectives, variables and objectives, as well as the dependencies learned between variables in other Bayesian network-based EDAs. This model leads to a problem decomposition that helps the proposed algorithm find better tradeoff solutions to the multiobjective problem. In addition to Pareto set approximation, the algorithm is also able to estimate the structure of the multiobjective problem. To apply the algorithm to many-objective problems, the algorithm includes four different ranking methods proposed in the literature for this purpose. The algorithm is first applied to the set of walking fish group problems, and its optimization performance is compared with a standard multiobjective evolutionary algorithm and another competitive multiobjective EDA. The experimental results show that on several of these problems, and for different objective space dimensions, the proposed algorithm performs significantly better and on some others achieves comparable results when compared with the other two algorithms. The algorithm is then tested on the set of CEC09 problems, where the results show that multiobjective optimization based on joint model estimation is able to obtain considerably better fronts for some of the problems compared with the search based on conventional genetic operators in the state-of-the-art multiobjective evolutionary algorithms. [ABSTRACT FROM AUTHOR]

The automatic classification of electrocardiogram (ECG) signals is of great clinical significance in eliminating the strenuous process of manually annotating ECG recordings. Although statistical models describing ECG signal dynamics currently exist, they depend considerably on a priori information and user-specified model parameters. Also, ECG beat morphologies, which vary greatly across different individuals and disease states, cannot easily be described by a single representation. In this paper, we propose sequential Bayesian based methods to effectively model and adaptively select parameters of ECG signals. We first consider an adaptive framework based on a sequential Bayesian tracking method that adaptively selects the best cardiac parameters by minimizing the estimation error and does not require early-stage processing to obtain prior signal information. We then present ECG modeling techniques using the interacting multiple model (IMM) and sequential Markov chain Monte Carlo (SMCMC) methods combined with simultaneous model selection. Both these methods can adaptively choose between different representations to model various ECG beat morphologies without requiring prior ECG information. The performance of the proposed algorithms is demonstrated using real ECG data. Finally, we develop a Bayesian maximum-likelihood based classifier to classify different types of cardiac arrhythmias using which, correction classification rates of 90% and 98% are obtained, when considering features obtained from the estimated model parameters of the adaptive framework, and both the IMM and SMCMC methods, respectively. [ABSTRACT FROM PUBLISHER]

Data generated from microarray experiments often suffer from missing values. As most downstream analyses need full matrices as input, these missing values have to be estimated. Bayesian principal component analysis (BPCA) is a well-known microarray missing value estimation method, but its performance is not satisfactory on datasets with strong local similarity structure. A bicluster-based BPCA (bi-BPCA) method is proposed in this paper to fully exploit local structure of the matrix. In a bicluster, the most correlated genes and experimental conditions with the missing entry are identified, and BPCA is conducted on these biclusters to estimate the missing values. An automatic parameter learning scheme is also developed to obtain optimal parameters. Experimental results on four real microarray matrices indicate that bi-BPCA obtains the lowest normalized root-mean-square error on 82.14% of all missing rates. [ABSTRACT FROM PUBLISHER]

In this paper, we present a formal model of human decision making in explore–exploit tasks using the context of multiarmed bandit problems, where the decision maker must choose among multiple options with uncertain rewards. We address the standard multiarmed bandit problem, the multiarmed bandit problem with transition costs, and the multiarmed bandit problem on graphs. We focus on the case of Gaussian rewards in a setting where the decision maker uses Bayesian inference to estimate the reward values. We model the decision maker's prior knowledge with the Bayesian prior on the mean reward. We develop the upper-credible-limit (UCL) algorithm for the standard multiarmed bandit problem and show that this deterministic algorithm achieves logarithmic cumulative expected regret, which is optimal performance for uninformative priors. We show how good priors and good assumptions on the correlation structure among arms can greatly enhance decision-making performance, even over short time horizons. We extend to the stochastic UCL algorithm and draw several connections to human decision-making behavior. We present empirical data from human experiments and show that human performance is efficiently captured by the stochastic UCL algorithm with appropriate parameters. For the multiarmed bandit problem with transition costs and the multiarmed bandit problem on graphs, we generalize the UCL algorithm to the block UCL algorithm and the graphical block UCL algorithm, respectively. We show that these algorithms also achieve logarithmic cumulative expected regret and require a sublogarithmic expected number of transitions among arms. We further illustrate the performance of these algorithms with numerical examples. [ABSTRACT FROM PUBLISHER]

In this paper, we present a new model for unsupervised discovery of recurrent temporal patterns (or motifs) in time series (or documents). The model is designed to handle the difficult case of multivariate time series obtained from a mixture of activities, that is, our observations are caused by the superposition of multiple phenomena occurring concurrently and with no synchronization. The model uses nonparametric Bayesian methods to describe both the motifs and their occurrences in documents. We derive an inference scheme to automatically and simultaneously recover the recurrent motifs (both their characteristics and number) and their occurrence instants in each document. The model is widely applicable and is illustrated on datasets coming from multiple modalities, mainly videos from static cameras and audio localization data. The rich semantic interpretation that the model offers can be leveraged in tasks such as event counting or for scene analysis. The approach is also used as a mean of doing soft camera calibration in a camera network. A thorough study of the model parameters is provided and a cross-platform implementation of the inference algorithm will be made publicly available. [ABSTRACT FROM AUTHOR]

The benefits of compressive sensing (CS) on counting and localization in wireless sensor networks have already been demonstrated in previous works. However, most existing solutions usually become infeasible in the presence of off-grid targets, which is inevitable for the traditional CS-based localization methods. To mitigate the error caused by off-grid targets, in this paper, a more accurate sparse approximation model is applied where the true and unknown sparsifying dictionary is approximated with its first-order Taylor expansion around a known dictionary. Based on the model, the counting and localization problem is formulated as a sparse reconstruction problem that recovers two sparse signals with related supports. To solve the problem, we develop an iterative algorithm under a variational Bayesian inference framework. Moreover, in practice, some faulty prior information (e.g., coarse positions) is usually available. To incorporate the information, a three-level hierarchical prior model is introduced and imposed on the sparse signals to be recovered, where the supports of sparse signals are learned in the third level. Additionally, a grid pruning mechanism and matrix inversion lemma are leveraged to accelerate the inference process. Extensive simulation results demonstrate that, compared with state-of-the-art algorithms, the proposed algorithm can achieve counting and localization with higher accuracy as well as lower cost. [ABSTRACT FROM PUBLISHER]

Conventional direction-of-arrival (DOA) estimation methods are sensitive to outlier measurements. Therefore, their performance may degrade substantially in the presence of impulsive noise. In this paper, we address the problem of DOA estimation in additive outliers from the perspective of sparse Bayesian learning (SBL). A Bayes-optimal algorithm is devised for robust DOA estimation, which can achieve excellent performance in terms of resolution and accuracy. To reduce the computational complexity of the SBL scheme, a fast alternating algorithm is also developed. New grid-refining procedures are further introduced into these two proposed algorithms to efficiently fix the off-grid gap. As our solutions do not require the prior knowledge of the number of sources and can resolve highly correlated or coherent sources, it is expected that they have higher applicability. Simulation results verify the outlier-robust performance of the SBL approach. [ABSTRACT FROM AUTHOR]

Situation-aware technologies enabled by multitarget tracking will lead to new services and applications in fields such as autonomous driving, indoor localization, robotic networks, and crowd counting. In this tutorial paper, we advocate a recently proposed paradigm for scalable multitarget tracking that is based on message passing or, more concretely, the loopy sum–product algorithm. This approach has advantages regarding estimation accuracy, computational complexity, and implementation flexibility. Most importantly, it provides a highly effective, efficient, and scalable solution to the probabilistic data association problem, a major challenge in multitarget tracking. This fact makes it attractive for emerging applications requiring real-time operation on resource-limited devices. In addition, the message passing approach is intuitively appealing and suited to nonlinear and non-Gaussian models. We present message-passing-based multitarget tracking methods for single-sensor and multiple-sensor scenarios, and for a known and unknown number of targets. The presented methods can cope with clutter, missed detections, and an unknown association between targets and measurements. We also discuss the integration of message-passing-based probabilistic data association into existing multitarget tracking methods. The superior performance, low complexity, and attractive scaling properties of the presented methods are verified numerically. In addition to simulated data, we use measured data captured by two radar stations with overlapping fields-of-view observing a large number of targets simultaneously. [ABSTRACT FROM PUBLISHER]

The sparse Bayesian learning based relevance vector machine (SBLRVM) algorithm is a promising algorithm to estimate the directions-of-arrival (DOAs) of multiple narrowband signals. The parameters involved in the DOA estimation model are automatically estimated by the algorithm that makes it more attractive than the deterministic sparsity based DOA estimation algorithms in which fine-tuning of parameters is necessary. However, one limitation of the algorithm is that it assumes the DOAs of the signals to be exactly aligned with the angular grids, which may not be true in practice. In this paper, we first propose an off-grid version of the narrowband SBLRVM algorithm. Next, we propose an off-grid wideband SBLRVM algorithm. The algorithms assume that the true scenario DOAs of the signals are not exactly aligned with the angular grids and the parameters of the algorithms are automatically estimated by the expectation maximization approach. In the wideband DOA estimation algorithm, we estimate one spatial power spectrum by simultaneously exploiting sparsity from all frequency bins. We demonstrate the application of the proposed algorithms by analyzing data from the shallow water HF\mathbf 97 ocean acoustic experiment. The estimated DOAs of a narrowband tonal from the experiment by using our proposed narrowband DOA estimation algorithm are consistent with the nonadaptive conventional beamformer. Processing a wideband chirp from the experiment shows that estimating one spatial power spectrum by simultaneously exploiting sparsity from all frequency bins using the proposed wideband DOA estimation algorithm is a more valuable processor than an incoherent combination of the power spectra from the individual frequency bins estimated using the proposed narrowband DOA estimation algorithm. Moreover, since our proposed algorithms are off-grid algorithms, an empirical analysis for the choice of the discretization interval of the angular spread is not required as opposed to the on-grid DOA estimation algorithms. This results in a reduced computational complexity. [ABSTRACT FROM PUBLISHER]

Moving object detection is about foreground and background separation based on motion detection. Detecting moving objects from similarly colored background (known as camouflage problem) has been a long-standing open question in this field. Discriminative modeling (DM), which focuses on enhancing the performance to distinguish foreground from background with discriminative features and well-designed classifiers, has been widely used for moving object detection. However, DM may tend to fail when encountering the camouflage problem, as the class separability in camouflaged areas is generally poor. In this paper, we propose a new strategy, camouflage modeling (CM), to identify camouflaged foreground pixels. In view of the fact that camouflage involves both foreground and background, we need to model both the background and the foreground, and compare them in a well-designed way in camouflage detection. Specifically, we develop a global model for the background, and an integration of global and local models for the foreground, respectively. Based on both background and foreground models, we introduce a factor to measure the degree of camouflage, and further identify truly camouflaged areas. In view of the fact that a moving object is usually composed of both camouflaged and noncamouflaged areas, CM and DM are fused in a Bayesian framework to perform complete object detection. Experiments are conducted on testing sequences to demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM PUBLISHER]

Swallowing accelerometry is a promising noninvasive approach for the detection of swallowing difficulties. In this paper, we propose an approach for classification of swallowing accelerometry recordings containing either healthy swallows or penetration–aspiration (entry of material into the airway) in dysphagic patients. The proposed algorithm is based on the wavelet packet decomposition of swallowing accelerometry signals in combination with linear discriminant analysis as a feature reduction method and Bayes classification. The proposed algorithm was tested using swallowing accelerometry signals collected from 40 patients during the regularly scheduled videoflouroscopy exam. The participants were instructed to swallow several 5-mL sips of thin liquid barium in a head neutral position. The results of our numerical analysis showed that the proposed algorithm can differentiate healthy swallows from aspiration swallows with an accuracy greater than 90%. These results position swallowing accelerometry as a valid approach for the detection of swallowing difficulties, particularly penetration–aspiration in patients suspected of dysphagia. [ABSTRACT FROM PUBLISHER]

Compressed sensing (CS) is on recovery of high dimensional signals from their low dimensional linear measurements under a sparsity prior and digital quantization of the measurement data is inevitable in practical implementation of CS algorithms. In the existing literature, the quantization error is modeled typically as additive noise and the multi-bit and 1-bit quantized CS problems are dealt with separately using different treatments and procedures. In this paper, a novel variational Bayesian inference based CS algorithm is presented, which unifies the multi- and 1-bit CS processing and is applicable to various cases of noiseless/noisy environment and unsaturated/saturated quantizer. By decoupling the quantization error from the measurement noise, the quantization error is modeled as a random variable and estimated jointly with the signal being recovered. Such a novel characterization of the quantization error results in superior performance of the algorithm which is demonstrated by extensive simulations in comparison with state-of-the-art methods for both multi-bit and 1-bit CS problems. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a Bayesian-inference-based recommendation system for online social networks. In our system, users share their content ratings with friends. The rating similarity between a pair of friends is measured by a set of conditional probabilities derived from their mutual rating history. A user propagates a content rating query along the social network to his direct and indirect friends. Based on the query responses, a Bayesian network is constructed to infer the rating of the querying user. We develop distributed protocols that can be easily implemented in online social networks. We further propose to use Prior distribution to cope with cold start and rating sparseness. The proposed algorithm is evaluated using two different online rating data sets of real users. We show that the proposed Bayesian-inference-based recommendation is better than the existing trust-based recommendations and is comparable to Collaborative Filtering (CF) recommendation. It allows the flexible tradeoffs between recommendation quality and recommendation quantity. We further show that informative Prior distribution is indeed helpful to overcome cold start and rating sparseness. [ABSTRACT FROM PUBLISHER]

Fault detection has been recognized in the semiconductor industry as an effective component of advanced process control framework in increasing yield and product quality. Principal component analysis (PCA) has been applied widely to semiconductor manufacturing process monitoring. However, the unique characteristics of semiconductor processes—high dimension of data, nonlinearity in most batch processes, and multimodal batch trajectories due to multiple operating conditions—significantly limit applicability of PCA to semiconductor manufacturing. This paper proposes a manifold learning algorithm, local and nonlocal preserving projection (LNPP), for feature extraction. Different from PCA, which aims to discover the global structure of Euclidean space, LNPP can find a good linear embedding that preserves local and nonlocal information. This may enable LNPP to find meaningful low-dimensional information hidden in high-dimensional observations. The Gaussian mixture model (GMM) is applied to handle process data with nonlinearity or multimodal features. GMM-based Mahalanobis distance is proposed to assess process states, and a Bayesian inference-based method is proposed to provide the process failure probability. A variable replacing-based contribution analysis method is developed to identify the process variables that are responsible for the onset of process fault. The proposed monitoring model is demonstrated through its application to a batch semiconductor etch process. [ABSTRACT FROM PUBLISHER]