For the problem of tracking maritime dim targets, the sequential Monte–Carlo multi-Bernoulli track-before-detect (SMC-MB-TBD) method is popular. However, this method may face low tracking accuracy and tracking loss due to particle impoverishment and velocity uncertainty. In this study, a novel filter called position scaling and velocity correction multi-Bernoulli (PSVC-MB) is proposed to deal with this problem. First, particle position scaling is used to replace resampling in the SMC-MB-TBD method to deal with the lack of particle diversity. Second, when the target is stably tracked, the target velocity is extracted from the multi-frame information and used for re-estimation. Pseudo point measurements are calculated from the weighted average of all locations near the particle position, and the particle velocity will be continuously corrected with the pseudo point measurements. Simulation results verify the effectiveness of the proposed method at different low signal-to-clutter ratios (SCRs).

To address the problem of underwater multi-sensor multi-target passive tracking in clutter, a distributed kernel mean embedding-based Gaussian belief propagation (DKME-GaBP) algorithm is proposed. First, a joint posterior probability density function (PDF) is established and factorized, and it is represented by the corresponding factor graph. Then, the GaBP algorithm is executed on this factor graph to reduce the computational complexity of data association. The factor graph of the GaBP consists of inner and outer loops. The inner loop is responsible for local track estimation and data association. The outer loop fuses information from different sensors. For the inner loop, the kernel mean embedding (KME) with a Gaussian kernel is designed to transform the strong nonlinear problem of local estimation into a linear problem in a high-dimensional reproducing kernel Hilbert space (RKHS). For the outer loop, a multi-sensor distributed fusion method based on KME is proposed to improve fusion accuracy by accounting for the distance among different PDFs in RKHS. The effectiveness and robustness of the DKME-GaBP are validated in the simulations.

The quotient of two multivariate Gaussian densities can be written as an unnormalized Gaussian density, which has been applied in some recently developed multiple-model fixed-interval smoothing algorithms. However, this expression is invalid if instead of being positive definite, the covariance of the unnormalized Gaussian density is indefinite (i.e., it has both positive and negative eigenvalues) or undefined (i.e., computing it requires inverting a singular matrix). This paper considers approximating the quotient of two Gaussian densities in this case using two different approaches to mitigate the caused numerical problems. The first approach directly replaces the indefinite covariance of the unnormalized Gaussian density with a positive definite matrix nearest to it. The second approach computes the approximation through solving, using the natural gradient, an optimization problem with a Kullback-Leibler divergence-based cost function. This paper illustrates the application of the theoretical results by incorporating them into an existing smoothing method for jump Markov systems and utilizing the obtained smoothers to track a maneuvering target.

This paper presents an enhanced two-step method for tracking an indoor point target using the time difference of arrival (TDOA) measurements from an ultra wideband (UWB) positioning system. Again, the algorithm preprocesses the raw TDOAs and then feeds the results to a recursively bounded grid-based filter (RBGF) for position tracking. Different from the state-of-the-art, inequality constraints on the true TDOAs from the RBGF are exploited in the preprocessing step through constrained Kullback-Leibler divergence (KLD) minimization. In particular, a semidefinite programming (S-DP) problem is formulated and solved to find a Gaussian TDOA posterior closest in terms of KLD to the unconstrained one while satisfying all inequality constraints. Simulations show that the newly developed algorithm outperforms the one we recently proposed to impose the inequality constraints via probability density function (PDF) truncation.

Click-through rate (CTR) prediction is a critical problem in web search, recommendation systems and online advertisement displaying. Learning good feature interactions is essential to reflect user's preferences to items. Many CTR prediction models based on deep learning have been proposed, but researchers usually only pay attention to whether state-of-the-art performance is achieved, and ignore whether the entire framework is reasonable. In this work, we use the discrete choice model in economics to redefine the CTR prediction problem, and propose a general neural network framework built on self-attention mechanism. It is found that most existing CTR prediction models align with our proposed general framework. We also examine the expressive power and model complexity of our proposed framework, along with potential extensions to some existing models. And finally we demonstrate and verify our insights through some experimental results on public datasets.

The Schmidt-Kalman filter (SKF) achieves filtering consistency in the presence of biases in system dynamic and measurement models through accounting for their impacts when updating the state estimate and covariance. However, the performance of the SKF may break down when the measurements are subject to non-Gaussian and heavy-tail noise. To address this, we impose the Wishart prior distribution on the precision matrix of measurement noise, such that the measurement likelihood now has heavier tails than the Gaussian distribution to deal with the potential occurrence of outliers. Variational inference is invoked to establish analytically tractable methods for computing the posterior of the system state, system biases, and the measurement noise precision matrix. The principle of the SKF considers the effect of system biases but does not actively estimate them when two variants of outlier-robust SKFs are incorporated. We evaluate their performance in terms of estimation accuracy and filtering consistency using simulations and real-world data. Promising results are obtained.

This paper considers exploiting the time difference of arrival (TDOA) measurements from a ultra wideband (UWB) indoor positioning system to locate a moving point target. In indoor environments, measured TDOAs are subject to large errors due to multipath and/or non-line-of-sight (NLOS) propagation. Besides, they are nonlinearly related to the target position. This paper presents an enhanced two-step approach to achieve robust TDOA indoor tracking. Similar to the existing method, the first-step of the new algorithm preprocesses the raw TDOAs to mitigate the effect of large TDOA errors while its second step applies a recursively bounded grid-based filter (RBGF) to achieve target position tracking. To improve performance, in this work, the possible target position area, which is explicitly obtained by the RGBF, is fed back to the first-step such that a constrained TDOA measurement preprocessing is now performed. Extensive simulation results show that the newly proposed scheme offers better TDOA estimation and indoor target positioning accuracy over the original method and other benchmark algorithms.

Distributed deep learning systems (DDLS) train deep neural network models by utilizing the distributed resources of a cluster. Developers of DDLS are required to make many decisions to process their particular workloads in their chosen environment efficiently. The advent of GPU-based deep learning, the ever-increasing size of datasets, and deep neural network models, in combination with the bandwidth constraints that exist in cluster environments require developers of DDLS to be innovative in order to train high-quality models quickly. Comparing DDLS side-by-side is difficult due to their extensive feature lists and architectural deviations. We aim to shine some light on the fundamental principles that are at work when training deep neural networks in a cluster of independent machines by analyzing the general properties associated with training deep learning models and how such workloads can be distributed in a cluster to achieve collaborative model training. Thereby we provide an overview of the different techniques that are used by contemporary DDLS and discuss their influence and implications on the training process. To conceptualize and compare DDLS, we group different techniques into categories, thus establishing a taxonomy of distributed deep learning systems.

This paper is inspired by how cognitive control manifests itself in the human brain and does so in a remarkable way. It addresses the many facets involved in the control of directed information flow in a dynamic system, culminating in the notion of information gap, defined as the difference between relevant information (useful part of what is extracted from the incoming measurements) and sufficient information representing the information needed for achieving minimal risk. The notion of information gap leads naturally to how cognitive control can itself be defined. Then, another important idea is described, namely the two-state model, in which one is the system's state and the other is the entropic state that provides an essential metric for quantifying the information gap. The entropic state is computed in the perceptual part (i.e., perceptor) of the dynamic system and sent to the controller directly as feedback information. This feedback information provides the cognitive controller the information needed about the environment and the system to bring reinforcement leaning into play; reinforcement learning (RL), incorporating planning as an integral part, is at the very heart of cognitive control. The stage is now set for a computational experiment, involving cognitive radar wherein the cognitive controller is enabled to control the receiver via the environment. The experiment demonstrates how RL provides the mechanism for improved utilization of computational resources, and yet is able to deliver good performance through the use of planning. The paper finishes with concluding remarks.

In this paper, we describe a cognitive radar (CR) that mimics the visual brain. Although the visual brain and radar are different in that the visual brain does not transmit a probing signal to the environment while the active radar greatly relies on the probing signal it transmits to the environment, both of them are observers of the surrounding environment. As such, there is much that we can learn from the visual brain in building a new generation of CRs that outperform traditional radars. In this paper, we confine the discussion, in both analytic and experimental terms, to CR aimed at target tracking. From a theoretical perspective, using the posterior Cramér-Rao lower bound (PCRLB), it is shown that a cognitive tracking radar has the potential to improve tracking performance significantly. In particular, computer experiments are presented, which demonstrate that CR can indeed go beyond the theoretical limits of traditional active radars (TARs) as well as fore-active radars (FARs); the latter are radars equipped with feedback from the receiver to the transmitter. Moreover, computer experiments are presented to demonstrate another practical benefit resulting from the combined use of memory and executive attention in CR for a target-tracking application. Specifically, it is shown that with the provision of these two cognitive processes, the transition in switching from one transmit waveform to another goes forward in a smooth manner. Such a capability is beyond that of TAR or FAR.

Subband-based eigenstructure method for bearing estimation (Xue, Wang and Liu (2004), Xue and Wang (2004)) require a mapping back manipulation after the subband frequency is estimated. In this paper, we propose a non-mapping back method to estimate the fullband bearings in the subband domain. The proposed method enhances the signal energy and reduces the root mean square error (RMSE). Simulations results show the decorrelation capability of coherent sources with appropriate subband division.

Wavelet packets-based MUSIC (WP-MUSIC) is proposed to improve the performance of classical MUSIC in scenarios of closely spaced DOA and low signal-to-noise ratio (SNR). With WP-MUSIC the fullband signal is decomposed into several subbands by wavelet packets, and then MUSIC is applied to each subband. The computational savings of WP-MUSIC compared to MUSIC are proven. Some simulation results, proving the validity and improved performance of the proposed approach, are presented.