Face clustering is crucial for applications such as facial recognition, identity verification, and surveillance in social systems. However, it is often hindered by noise and false-positive connections at cluster boundaries, which significantly degrade performance. To address these challenges, we propose a novel face clustering framework with neighbor structure refinement (NSR-FC), which enhances clustering effectiveness by refining neighbor structures across three dimensions: features, density, and connectivity. Within NSR-FC, cascaded graph convolutional networks (C-GCN) improve feature extraction while simultaneously optimizing the graph structure to reduce noise and generate discriminative features. Leveraging these refined features, we construct a local density graph using updated affinities from the k-nearest neighbor (KNN) graph, effectively eliminating negative pairs while preserving positive ones. Furthermore, we assess edge connectivity within the density graph to construct a global connectivity graph. Finally, the clustering results are obtained by applying breadth-first search (BFS) to the union graph edge set. Extensive experiments on benchmark datasets such as MS-Celeb-1M demonstrate that NSR-FC achieves state-of-the-art (SOTA) performance, underscoring its effectiveness in advancing face clustering.

In multi-label learning, traditional methods try to directly establish mapping functions between samples and their labels. However, such methods may suffer low classification performance due to inherent noise and incoherent representation of samples. Therefore, considering the correlation between samples and their neighbors, as they may share common feature semantics, a multi-label classification method via feature enhancement from neighborhood samples, referred to as DRNSA is proposed. In this method, we construct a Laplacian graph dynamically, by considering the distance of two samples in a projected low rank subspace. With this neighborhood selection strategy, we then build a multi-label classifier via feature enhancement from neighborhood samples. Different from past works that directly build classifiers from samples and their labels, we build a new enhanced data sample which is weighted by its semantically similar neighborhood samples. Thus, our method can obtain better robust subspace in very high noisy data representations. Experiments conducted on cal500, corel5k, corel16k1 and corel16k4 datasets, show a significant out-performance of our DRNSA method over multi-label classification methods, including MLSF, MLFE, BR, PLST, CSSP, CPLST, and FaIE.

Deep Compressed Sensing (DCS) has attracted considerable interest due to its superior quality and speed compared to traditional CS algorithms. However, current approaches employ simplistic convolutional downsampling to acquire measurements, making it difficult to retain high-level features of the original signal for better image reconstruction. Furthermore, these approaches often overlook the presence of both high- and low-frequency information within the network, despite their critical role in achieving high-quality reconstruction. To address these challenges, we propose a novel Multi-Cross Sampling and Frequency Division Network (MCFD-Net) for image CS. The Dynamic Multi-Cross Sampling (DMCS) module, a sampling network of MCFD-Net, incorporates pyramid cross convolution and dual-branch sampling with multi-level pooling. Additionally, it introduces an attention mechanism between perception blocks to enhance adaptive learning effects. In the second deep reconstruction stage, we design a Frequency Division Reconstruction Module (FDRM). This module employs a discrete wavelet transform to extract high- and low-frequency information from images. It then applies multi-scale convolution and self-similarity attention compensation separately to both types of information before merging the output reconstruction results. The MCFD-Net integrates the DMCS and FDRM to construct an end-to-end learning network. Extensive CS experiments conducted on multiple benchmark datasets demonstrate that our MCFD-Net outperforms state-of-the-art approaches, while also exhibiting superior noise robustness.

The previous relation-based knowledge distillation methods tend to construct global similarity relationship matrix in a mini-batch while ignoring the knowledge of neighbourhood relationship. In this paper, we propose a new similarity-based relational knowledge distillation method that transfers neighbourhood relationship knowledge by selecting K-nearest neighbours for each sample. Our method consists of two components: Neighbourhood Feature Relationship Distillation and Neighbourhood Logits Relationship Distillation. We perform extensive experiments on CIFAR100 and Tiny ImageNet classification datasets and show that our method outperforms the state-of-the-art knowledge distillation methods. Our code is available at: https://github.com/xinxiaoxiaomeng/NRKD.git.

Abstract Website fingerprinting (WF) attacks based on deep neural networks pose a significant threat to the privacy of anonymous network users. However, training a deep WF model requires many labeled traces, which can be labor-intensive and time-consuming, and models trained on the originally collected traces cannot be directly used for the classification of newly collected traces due to the concept drift caused by the time gap in the data collection. Few-shot WF attacks are proposed for using the originally and few-shot newly collected labeled traces to facilitate anonymous trace classification. However, existing few-shot WF attacks ignore the fine-grained feature alignment to eliminate the concept drift in the model training, which fails to fully use the knowledge of labeled traces. We propose a novel few-shot WF attack called Joint Alignment Networks (JAN), which conducts fine-grained feature alignment at both semantic-level and feature-level. Specifically, JAN minimizes a distribution distance between originally and newly collected traces in the feature space for feature-level alignment, and utilizes two task-specific classifiers to detect unaligned traces and force these traces mapped within decision boundaries for semantic-level alignment. Extensive experiments on public datasets show that JAN outperforms the state-of-the-art few-shot WF methods, especially in the difficult 1-shot tasks.

Density peak clustering (DPC) has garnered growing interest over recent decades due to its capability to identify clusters with diverse shapes and its resilience to the presence of noisy data. Most DPC-based methods exhibit high computational complexity. One approach to mitigate this issue involves utilizing density subgraphs. Nevertheless, the utilization of density subgraphs may impose restrictions on cluster sizes and potentially lead to an excessive number of small clusters. Furthermore, effectively handling these small clusters, whether through merging or separation, to derive accurate results poses a significant challenge, particularly in scenarios where the number of clusters is unknown. To address these challenges, we propose an adaptive density subgraph clustering algorithm (ADSC). ADSC follows a systematic three-step procedure. First, the highdensity regions in the dataset are recognized as density subgraphs based on k-nearest neighbor (KNN) density. Second, the initial clustering is carried out by utilizing an automated mechanism to identify the important density subgraphs and allocate outliers. Last, the obtained initial clustering results are further refined in an adaptive manner using the cluster self-ensemble technique, ultimately yielding the final clustering outcomes. The clustering performance of the proposed ADSC algorithm is evaluated on nineteen benchmark datasets. The experimental results demonstrate that ADSC possesses the ability to automatically determine the optimal number of clusters from intricate density data, all while maintaining high clustering efficiency. Comparative analysis against other well-known density clustering algorithms that require prior knowledge of cluster numbers reveals that ADSC consistently achieves comparable or superior clustering results.

The finite inverted beta mixture model (IBMM) has been proven to be efficient in modeling positive vectors. Under the traditional variational inference framework, the critical challenge in Bayesian estimation of the IBMM is that the computational cost of performing inference with large datasets is prohibitively expensive, which often limits the use of Bayesian approaches to small datasets. An efficient alternative provided by the recently proposed stochastic variational inference (SVI) framework allows for efficient inference on large datasets. Nevertheless, when using the SVI framework to address the non-Gaussian statistical models, the evidence lower bound (ELBO) cannot be explicitly calculated due to the intractable moment computation. Therefore, the algorithm under the SVI framework cannot directly use stochastic optimization to optimize the ELBO, and an analytically tractable solution cannot be derived. To address this problem, we propose an extended version of the SVI framework with more flexibility, namely, the extended SVI (ESVI) framework. This framework can be used in many non-Gaussian statistical models. First, some approximation strategies are applied to further lower the ELBO to avoid intractable moment calculations. Then, stochastic optimization with noisy natural gradients is used to optimize the lower bound. The excellent performance and effectiveness of the proposed method are verified in real data evaluation.

Abstract Previous studies have demonstrated the usefulness of employing automated static analysis tools (ASAT) and techniques to detect security bugs in software systems. However, these studies are usually focused on analyzing the effectiveness of the tools using open‐source tools based on C/C++ source code. The choice for making an appropriate decision on the most suitable tool for bug detection in Java code software remains a relatively unexplored domain. To address this deficiency, this study empirically evaluates eight widely used ASATs, namely, Findbug, PMD, YASCA, LAPSE+, JLint, Bandera, ESC/Java, and Java Pathfinder using the Juliet Test Suite (Test Suite v1.2). Additionally, we assessed the performance of the detection capabilities for the aforementioned bug detection tools using robust performance measures such as precision, recall, Youden index, and the OWASP web benchmark evaluation (WBE). The experimental results show that the tools obtain precision values ranging from 83% to 90.7% based on the studied datasets. Specifically, the Java Pathfinder achieves the best precision score of 90.7%, followed by YASCA and Bandera with a precision score of 88.7% and 83%, respectively. Similarly, Bandera, ESC/Java, and Java Pathfinder obtain a Youden index of 0.8, which indicates the effectiveness of the tools in detecting security bugs in Java source code.

Bayesian estimation of parameters in the Dirichlet mixture process of the Beta-Liouville distribution (i.e., the infinite Beta-Liouville mixture model) has recently gained considerable attention due to its modeling capability for proportional data. However, applying the conventional variational inference (VI) framework cannot derive an analytically tractable solution since the variational objective function cannot be explicitly calculated. In this paper, we adopt the recently proposed extended VI framework to derive the closed-form solution by further lower bounding the original variational objective function in the VI framework. This method is capable of simultaneously determining the model's complexity and estimating the model's parameters. Moreover, due to the nature of Bayesian nonparametric approaches, it can also avoid the problems of underfitting and overfitting. Extensive experiments were conducted on both synthetic and real data, generated from two real-world challenging applications, namely, object detection and text categorization, and its superior performance and effectiveness of the proposed method have been demonstrated.

The application of cross-corpus for speech emotion recognition (SER) via domain adaptation methods have gain high acknowledgment for developing good robust emotion recognition systems using different corpora or datasets. However, the issue of cross-lingual still remains a challenge in SER and needs more attention to resolve the scenario of applying different language types in both training and testing. In this paper, we propose a triple attentive asymmetric convolutional neural network to address the recognition of emotions for cross-lingual and cross-corpus speech in an unsupervised approach. The proposed method adopts the joint supervision of softmax loss and center loss to learn high power discriminative feature representations for target domain via the use of high quality pseudo-labels. The proposed model uses three attentive convolutional neural networks asymmetrically, where two of the networks are used to artificially label unlabeled target samples as a result of their predictions from training on source labeled samples and the other network is used to obtain salient target discriminative features from the pseudo-labeled target samples. We evaluate our proposed method on three different language types (i.e., English, German, and Italian) data sets. The experimental results indicate that, our proposed method achieves higher prediction accuracy over other state-of-the-art methods.

Facial expression recognition (FER) in the wild is a very challenging problem due to different expressions under complex scenario (e.g., large head pose, illumination variation, occlusions, etc.), leading to suboptimal FER performance. Accuracy in FER heavily relies on discovering superior discriminative, emotion-related features. In this paper, we propose an end-to-end module to disentangle latent emotion discriminative factors from the complex factors variables for FER to obtain salient emotion features. The training of proposed method contains two stages. First of all, emotion samples are used to obtain the latent representation using a variational auto-encoder with reconstruction penalization. Furthermore, the latent representation as the input is thrown into a disentangling layer to learn a set of discriminative emotion factors through the attention mechanism (e.g., a Squeeze-and-Excitation block) that encourages to separate emotion-related factors and nonaffective factors. Experimental results on public benchmark databases (RAF-DB and FER2013) show that our approach has remarkable performance in complex scenes than current state-of-the-art methods.

Finding an effective method for testing object-oriented software (OOS) has proven elusive in the software community due to the rapid development of object-oriented programming (OOP) technology. Although significant progress has been made by previous studies, challenges still exist in relation to the object distance measurement of OOS using Adaptive Random Testing (ART). This is partly due to the unique features of OOS such as encapsulation, inheritance and polymorphism. In a previous work, we proposed a new similarity metric called the Object and Method Invocation Sequence Similarity (OMISS) metric to facilitate multi-class level testing using ART. In this paper, we broaden the set of models in the metric (OMISS) by considering the method parameter and adding the weight in the metric to develop a new distance metric to improve unit testing of OOS. We used the new distance metric to calculate the distance between the set of objects and the distance between the method sequences of the test cases. Additionally, we integrate the new metric in unit testing with ART and applied it to six open source subject programs. The experimental result shows that the proposed method with method parameter considered in this study is better than previous methods without the method parameter in the case of the single method. Our finding further shows that the proposed unit testing approach is a promising direction for assisting software engineers who seek to improve the failure-detection effectiveness of OOS testing.

Linear reconstruction measure (LRM) is a promising similarity measure of data. In this paper, we consider the locality of data in LRM, and propose weighted two-phase linear reconstruction measure-based classification (WTPLRMC). In WTPLRMC, the first phase determines the representative training samples from all training samples by LRM, and the second phase constrains the linear reconstruction coefficients of the chosen representative training samples in first phase using the locality of data, which is reflected by the similarity weights between each test sample and the representative training samples. The effectiveness of the proposed WTPLRMC is well demonstrated on some benchmark face databases with satisfactory classification results.

A key objective of software testing is to find program errors that cause failure in software, at less cost. One basic testing technique is random testing (RT), but many researchers have criticised its failure‐detection effectiveness. Several researchers have proposed that an enhancement of the failure‐detection effectiveness of RT is achieved if test cases are evenly spread within the input domain. Adaptive RT (ART) describes a family of algorithms that employ various strategies to evenly and randomly spread test cases. Fixed sized candidate set ART (FSCS‐ART) is an ART algorithm that has gained many research studies far and wide; however, the high distance computations make its algorithm computationally expensive. The authors propose a new ART method that restricts distance computations to only test cases inside an exclusion zone. The experimental results show that the new ART method not only improves RT but also provides failure‐detection effectiveness similar to FSCS‐ART, while significantly minimising computation overhead.

Device-free motion tracking with radio tomographic networks using received signal strength (RSS) measurements has attracted considerable research efforts. Since the motion scene to be reconstructed can often be assumed sparse, i.e., it consists only of several targets, the Compressed Sensing (CS) framework can be applied. We cast the motion tracking as a CS problem and employ an efficient algorithm, Orthogonal Matching Pursuit (OMP), for sparse recovery. Furthermore, we exploit a feedback structure which leads to a substantial reduction of the amount of measurements. The feedback structure utilizes the prior knowledge (locations of targets) in time sequence to predict next frame support. Compared with the least-square type methods, the proposed motion tracking based on feedback sparse recovery can directly determine where the targets are located in the network area and reduce the amount of measurements required for reliable tracking. Experimental results show its favorable performance.