Breast cancer has become the most common cancer in the world, and biopsy is the most reliable and widely used technique for detecting breast cancer. However, observation of histopathological images is time-consuming and labor-intensive. Currently, CNN has become the mainstream method for breast cancer histopathological image classification research. However, some studies have found that the optical microscope-generated histopathological images have noise, and the output of a well-trained convolutional neural network in image classification tasks can change drastically due to small variations in the input. Therefore, the quality of the image significantly affects the accuracy of the classification. Wavelet transform is a commonly used denoising method, but the selection of the threshold is a difficult problem, and traditional methods are difficult to find the appropriate threshold quickly and accurately. This paper proposes an adaptive threshold selection method that combines threshold selection steps with deep learning methods by using the threshold as a parameter in the CNN model to train. In this way, we associate the threshold with the classification result of the model and find the appropriate value for that image and task by back-propagation in training. The method was experimented on publicly available datasets BreaKHis and BACH. The results in BreaKHis (40x: 94.37 % , 100x: 93.85 % , 200x: 91.63 % , 400x: 93.31 % ), and BACH (91.25 % ) demonstrate that our adaptive threshold selection method can improve classification accuracy and is significantly superior to traditional threshold selection methods.

Model extraction attacks (MEAs) allow adversaries to replicate a surrogate model analogous to the target model's decision pattern. While several attacks and defenses have been studied in-depth, the underlying reasons behind our susceptibility to them often remain unclear. Analyzing these implication influence factors helps to promote secure deep learning (DL) systems, it requires studying extraction attacks in various scenarios to determine the success of different attacks and the hallmarks of DLs. However, understanding, implementing, and evaluating even a single attack requires extremely high technical effort, making it impractical to study the vast number of unique extraction attack scenarios. To this end, we present a first-of-its-kind holistic evaluation of implication factors for MEAs which relies on the attack process abstracted from state-of-the-art MEAs. Specifically, we concentrate on four perspectives. we consider the impact of the task accuracy, model architecture, and robustness of the target model on MEAs, as well as the impact of the model architecture of the surrogate model on MEAs. Our empirical evaluation includes an ablation study over sixteen model architectures and four image datasets. Surprisingly, our study shows that improving the robustness of the target model via adversarial training is more vulnerable to model extraction attacks.

One key factor able to boost the applications of artificial intelligence (AI) in security-sensitive domains is to leverage them responsibly, which is engaged in providing explanations for AI. To date, a plethora of explainable artificial intelligence (XAI) has been proposed to help users interpret model decisions. However, given its data-driven nature, the explanation itself is potentially susceptible to a high risk of exposing privacy. In this paper, we first show that the existing XAI is vulnerable to model extraction attacks and then present an XAI-aware dual-task model extraction attack (DTMEA). DTMEA can attack a target model with explanation services, that is, it can extract both the classification and explanation tasks of the target model. More specifically, the substitution model extracted by DTMEA is a multitask learning architecture, consisting of a sharing layer and two task-specific layers for classification and explanation. To reveal which explanation technologies are more vulnerable to expose privacy information, we conduct an empirical evaluation of four major explanation types in the benchmark data set. Experimental results show that the attack accuracy of DTMEA outperforms the predicted-only method with up to 1.25%, 1.53%, 9.25%, and 7.45% in MNIST, Fashion-MNIST, CIFAR-10, and CIFAR-100, respectively. By exposing the potential threats on explanation technologies, our research offers the insights to develop effective tools that are able to trade off security-sensitive relationships.

Deep neural networks (DNNs) have gained widespread adoption in computer vision. Unfortunately, state-of-the-art DNNs are vulnerable to adversarial example (AE) attacks, where an adversary introduces imperceptible perturbations to a test example for defrauding DNNs. The obstacles have urged intensive research on improving the DNN robustness via adversarial training, that is, the clean data set is blended with adversarial examples to carry out training. However, the adversarial example attack technologies are open-ended, and the adversarial training is insufficient to focus on improving robustness performance. To circumvent this limitation, we mitigate adversarial example attacks from another perspective, which aims at detecting adversarial examples. Feature autoencoder detector (FADetector), a novel defense framework that exploits feature knowledge is proposed. One of the hallmarks of FADetector is to not involve adversarial examples to train the detector. Our extensive evaluation on MNIST and CIFAR-10 data sets demonstrates that our defense outperforms the conventional autoencoder detectors in terms of detection accuracy.

Rumors mislead judgments of people, affect economic development, and the stability of social order. The research on the rule of spreading rumors is significant and meaningful. This paper improves the traditional Barabási–Albert scale-free network and proposes a network topology model that conforms to the characteristics of sharing social networks based on the complex network theory and the actual characteristics of sharing social networks. In addition, the credulous spider rational taciturn rumor propagation model is proposed by improving the credulous spider rational model, which solves the overspread problem of the traditional rumor propagation model. This paper further studies the influence of anxiety on the spread of rumors, and finds that the anxiety of audience is increasing with the spread degree of rumors.

Identification of drug–target interactions (DTIs) is critical for discovering potential target protein candidates for new drugs. However, traditional experimental methods have limitations in discovering DTIs. They are time-consuming, tedious, and expensive, and often suffer from high false-positive rates and false-negative rates. Therefore, using computational methods to predict DTIs has received extensive attention from many researchers in recent years. To address this issue, in this paper, an effective prediction model is presented which is based on the information of drug molecular structure data and protein sequence data. It performs prediction with the following procedures. First, we transform the sequences of each target into a position-specific scoring matrix (PSSM), such that the features can retain biological evolutionary information. We then use a feature vector of molecular substructure fingerprints to describe the chemical structure information of the drug compounds. Second, the Legendre moments algorithm is used to extract new features from the PSSM. Finally, a classification algorithm called rotation forest is used to perform prediction, we tested its prediction performance on four golden standard data sets: enzymes, G-protein-coupled receptors, ion channels, and nuclear receptors. As a result, the proposed method achieves average accuracies of 0.9026, 0.8260, 0.8703, and 0.7444 on these four data sets using five-fold cross-validation. We also compare the proposed method with the support vector machine and other existing approaches. The proposed model is proved to be superior to comparative methods, showing that it is feasible, effective, and robust for predicting potential DTI.

As machine learning (ML) technologies evolve, various online intelligent services use ML models to provide predictions. Unfortunately, attackers can obtain the private information of the model by interacting with the online service, namely model inversion attack (MIA). However, MIA requires large data sets to be transferred to an online service to obtain the predictive value of the inference model. Besides, the huge transmission may cause the administrator's active defense. To overcome this drawback, we propose a novel MIA scheme, which leverages latent information extracted by an auxiliary neural network as high-dimensional features to simplify what inversion model should learn. The core idea of our scheme is to reuse some parameters of the local pretraining model. Extensive experiments have verified the effectiveness of our method in convolutional neural networks on LFW, pubFig, MNIST data sets. Experimental results show that even with a few queries, our inversion method still work accurately and is superior to other technologies. It is worth mentioning that our method makes it more difficult for administrators to defend against the attack and elicit more investigations for privacy-preserving.

As we all know, video frame rate determines the quality of the video. The higher the frame rate, the smoother the movements in the picture, the clearer the information expressed, and the better the viewing experience for people. Video interpolation aims to increase the video frame rate by generating a new frame image using the relevant information between two consecutive frames, which is essential in the field of computer vision. The traditional motion compensation interpolation method will cause holes and overlaps in the reconstructed frame, and is easily affected by the quality of optical flow. Therefore, this paper proposes a video frame interpolation method via optical flow estimation with image inpainting. First, the optical flow between the input frames is estimated via combined local and global-total variation (CLG-TV) optical flow estimation model. Then, the intermediate frames are synthesized under the guidance of the optical flow. Finally, the nonlocal self-similarity between the video frames is used to solve the optimization problem, to fix the pixel loss area in the interpolated frame. Quantitative and qualitative experimental results show that this method can effectively improve the quality of optical flow estimation, generate realistic and smooth video frames, and effectively increase the video frame rate.

Perpetrators leverage the untraceable feature to conduct illegal behaviors leading security issues with respect to mixing coins. Generally, bad coins are blocked based on a common blacklist. However, the blacklist may not be updated in time, which results in that bad coins escape the blocking. Consequently, perpetrators can still conduct illicit behaviors such as money laundering. In this paper, we apply game theory under imperfect information to study how coins' quality restrain these illicit behaviors under the incomplete scenario. More specifically, we propose a strategy for participants to submit deposits if they hope to mix coins with others even if they are not in blacklist at this time. The deposits will not be refunded when participants are included in the blacklist after mixing. Therefore, no participants have incentives to mix with bad coins. At the last part of this paper, we also simulate the incomes for participants, which indicates that deposits strategy is effective to prevent illicit behaviors.

As is known, kernel methods are developed to handle nonlinear classification. However, we have found that based on vector representation of face images, it is not easy to improve the performance of LDA-based methods by incorporating the kernel trick. Actually, in the case that the dimensionality of face images in vector form is far greater than the number of samples, linear classifiers are adequate and it is illogical to increase dimensions of vectors using kernel methods. In order to give full play to the strong point of the kernel technique for manipulating the nonlinearity of pattern distribution, we propose a new image feature extraction method for face recognition, called kernel-based two-dimensional Fisher discriminant analysis (K2DFDA), which deals with a face image directly as a matrix, instead of a stacked vector from rows or columns of the image. Moreover, we present a kernel parameter optimization scheme for K2DFDA, based on the maximum margin criterion and the damped Newton's method. Experimental results show the effectiveness of K2DFDA and its parameter optimization scheme.

Kernel-based methods have been proved to be an effective approach for face recognition in dealing with complex and nonlinear face image variations. While many encouraging results have been reported, the selection of kernel is ratheradhoc. This letter proposes a systematic method to construct a new kernel for kernel discriminant analysis, which is good for handling illumination problem. The proposed method first learns a kernel matrix by maximizing the difference between inter-class and intra-class similarities under the Lambertian model, and then generalizes the kernel matrix to our proposed ILLUM kernel using the scattered data interpolation technique. Experiments on the Yale-B and the CMU PIE face databases show that, the proposed kernel outperforms the popular Gaussian kernel in Kernel Discriminant Analysis and the recognition rate can be improved around 10%.

This paper presents a new algorithm named Kernel Bisecting k-means and Sample Removal (KBK-SR) as a sampling preprocessing for SVM training to improve the scalability. The novel clustering approach Kernel Bisecting k-means in the KBK-SR tends to fast produce balanced clusters of similar sizes in the kernel feature space, which makes KBK-SR efficient and effective for reducing training samples for nonlinear SVMs. Theoretical analysis and experimental results on three UCI real data benchmarks both show that, with very short sampling time, our algorithm dramatically accelerates SVM training while maintaining high test accuracy.