Millimeter-wave radar has been widely adopted in intelligent transportation systems. Modern millimeter-wave radars' high resolution makes a single target yield multiple measurements (point cloud measurements), turning target tracking into an extended target tracking (ETT) problem. The uncertainty in radar measurement sources coupled with the complex spatial distribution of measurements fundamentally challenges ETT algorithms. Conventional ETT algorithms with cluster-then-associate frameworks partition point cloud measurements predominantly through density, failing to exploit target shape information and resulting in suboptimal clustering efficacy. This leads to inaccurate associations between measurements and extended targets, ultimately degrading overall tracking accuracy. This paper proposes a novel approach to circumvent the performance limitations caused by clustering errors in traditional methods. First, we propose a clustering-free closed-loop ETT framework that incorporates the prior target's shape information as feedback. Subsequently, we develop a data association method that leverages the inherent correlation between point cloud measurements and target shape. The associated measurements are then probabilistically fused and integrated into a Kalman filter for state updating. In simulated and real-world datasets, compared with the traditional method, clustering before association, we have validated the effectiveness of the proposed method.

The study of cells and their responses to genetic or chemical perturbations promises to accelerate the discovery of therapeutics targets. However, designing adequate and insightful models for such data is difficult because the response of a cell to perturbations essentially depends on contextual covariates (e.g., genetic background or type of the cell). There is therefore a need for models that can identify interactions between drugs and contextual covariates. This is crucial for discovering therapeutics targets, as such interactions may reveal drugs that affect certain cell types but not others. We tackle this problem with a novel Factorized Causal Representation (FCR) learning method, an identifiable deep generative model that reveals causal structure in single-cell perturbation data from several cell lines. FCR learns multiple cellular representations that are disentangled, comprised of covariate-specific (Z_x), treatment-specific (Z_t) and interaction-specific (Z_tx) representations. Based on recent advances of non-linear ICA theory, we prove the component-wise identifiability of Z_tx and block-wise identifiability of Z_t and Z_x. Then, we present our implementation of FCR, and empirically demonstrate that FCR outperforms state-of-the-art baselines in various tasks across four single-cell datasets.

Multi-frame detection has been widely researched in the scenario where the target signal-to-noise is low. However, it becomes a challenging problem under heterogeneous clutter environment. As strong clutter energy is accumulated along with the target in multiple frames, low SNR targets are still annihilated in clutter. To achieve effective clutter suppression and dim targets detection, a novel multi-frame procedure for energy accumulation under heterogeneous clutter is proposed in this paper. The presented architecture concerns a Space-Time Adaptive Processing (STAP) processor and a multi-frame detector. The STAP processor calculates the clutter covariance matrix using multi-frame training samples near the cell under test and extracts data contaminated by the target component. The multi-frame detector is developed to detect dim targets and output estimated target track sequences. Finally, simulation results are given to demonstrate the efficacy of the proposed algorithm.

Weak target recognition, tracking and track management with a low signal-to-noise ratio (SNR) are always tricky problems. Probability hypothesis density (PHD) filtering propagates the first-order multi-target moment to obtain the best Poisson approximation to multi-target density. The PHD filtering does not consider explicit associations between measurements and targets, which is computationally efficient. But it cannot distinguish different targets or extract the time series of track states. Based on track-before-detect (TBD) strategies, this paper proposes labeled PHD (LPHD) filtering and derives its close-form solution, which identifies targets with a unique label. It is derived based on rigorous Bayes criteria, finite set statistics and Kullback-Leibler divergence minimization approximation. The separable TBD-based observation likelihood is conjugate to the Poisson mixture prior for LPHD filtering. Under the point-target assumption, the multi-hypothesis assignments of pixel-to-target are implemented with Murty’s K-shortest path algorithm for LPHD filtering. Additionally, sequential Monte Carlo (SMC) implementations under the nonlinear non-Gaussian assumption are devised. Finally, simulations exhibit good performance in low SNR scenarios.

Single-cell RNA sequencing (scRNA-seq) experiments measure transcriptional profiles that encode diverse sources of variation, both biological and technical, and complex coordinations among them. PCA is a popular method for interpretable dimension reduction. However, it assumes a linear mapping between data and latent components and this may not be warranted in complex data. Single cell variational inference (scVI) offers a nonlinear method for latent space mapping, but the latent factors are not in general interpretable. In light of disentangled representations that learn independent data generative factors of single cell data in an unsupervised way, we propose factor variational inference (factorVI). FactorVI learns the disentangled factors among biologically relevant latent variables directly by penalizing correlations between them. We evaluate the factorVI through clustering in publicly available datasets and we observe high accuracy. We also propose biological interpretation of the latent factors.

Instance transfer approaches consider source and target data together during the training process, and borrow examples from the source domain to augment the training data, when there is limited or no label in the target domain. Among them, boosting-based transfer learning methods (e.g., TrAdaBoost) are most widely used. When dealing with more complex data, we may consider the more complex hypotheses (e.g., a decision tree with deeper layers). However, with the fixed and high complexity of the hypotheses, TrAdaBoost and its variants may face the overfitting problems. Even worse, in the transfer learning scenario, a decision tree with deep layers may overfit different distribution data in the source domain. In this paper, we propose a new instance transfer learning method, i.e., Deep Decision Tree Transfer Boosting (DTrBoost), whose weights are learned and assigned to base learners by minimizing the data-dependent learning bounds across both source and target domains in terms of the Rademacher complexities. This guarantees that we can learn decision trees with deep layers without overfitting. The theorem proof and experimental results indicate the effectiveness of our proposed method.

High-throughput technologies have enabled us to rapidly accumulate a wealth of diverse data types. These multi-view data contain much more information to uncover the cluster structure than single-view data, which draws raising attention in data mining and machine learning areas. On one hand, many features are extracted to provide enough information for better representations, on the other hand, such abundant features might result in noisy, redundant and irrelevant information, which harms the performance of the learning algorithms. In this paper, we focus on a new topic, multi-view unsupervised feature selection, which aims to discover the discriminative features in each view for better explanation and representation. Although there are some exploratory studies along this direction, most of them employ the traditional feature selection by putting the features in different views together and fail to evaluate the performance in the multi-view setting. The features selected in this way are difficult to explain due to the meaning of different views, which disobeys the goal of feature selection as well. In light of this, we intend to give a correct understanding of multi-view feature selection. Different from the existing work, which either incorrectly concatenates the features from different views, or takes huge time complexity to learn the pseudo labels, we propose a novel algorithm, Robust Multi-view Feature Selection (RMFS), which applies robust multi-view K-means to obtain the robust and high quality pseudo labels for sparse feature selection in an efficient way. Nontrivially we give the solution by taking the derivatives and further provide a K-means-like optimization to update several variables in a unified framework with the convergence guarantee. We demonstrate extensive experiments on three real-world multi-view data sets, which illustrate the effectiveness and efficiency of RMFS in terms of both single-view and multi-view evaluations by a large margin.

The problem of super resolution of partial pixelated images is considered in this paper. Partial pixelated images are more and more common in nowadays due to public safety etc. However, in some special cases, for instance criminal investigation, some images are pixelated intentionally by criminals and partial pixelate make it hard to reconstruct images even a higher resolution images. Hence, a method is proposed to handle this problem based on the deep convolutional neural network, termed depixelate super resolution CNN(DSRCNN). Given the mathematical expression pixelates, we propose a model to reconstruct the image from the pixelation and map to a higher resolution by combining the adversarial autoencoder with two depixelate layers. This model is evaluated on standard public datasets in which images are pixelated randomly and compared to the state of arts methods, shows very exciting performance.