Search

With the gradual popularization of autonomous driving technology, how to obtain traffic sign information efficiently and accurately is very important for subsequent decision-making and planning tasks. Traffic sign detection and recognition (TSDR) algorithms include color-based, shape-based, and machine learning based. However, the algorithms mentioned above are insufficient for traffic sign detection tasks in complex environments. In this paper, we propose a traffic sign detection and recognition paradigm based on deep learning algorithms. First, to solve the problem of insufficient spatial information in high-level features of small traffic signs, the parallel deformable convolution module (PDCM) is proposed in this paper. PDCM adaptively acquires the corresponding receptive field preserving the integrity of the abstract information through symmetrical branches thereby improving the feature extraction capability. Simultaneously, we propose sub-pixel convolution attention module (SCAM) based on the attention mechanism to alleviate the influence of scale distribution. Distinguishing itself from other feature fusion, our proposed method can better focus on the information of scale distribution through the attention module. Eventually, we introduce GSConv to further reduce the computational complexity of our proposed algorithm, better satisfying industrial application. Experimental results demonstrate that our proposed methods can effectively improve performance, both in detection accuracy and mAP@0.5. Specifically, when the proposed PDCM, SCAM, and GSConv are applied to the Yolov5, it achieves 89.2% mAP@0.5 in TT100K, which exceeds the benchmark network by 4.9%. [ABSTRACT FROM AUTHOR]

In the development of algorithms for convex optimization problems, symmetry plays a very important role in the approximation of solutions in various real-world problems. In this paper, based on a fixed point algorithm with the inertial technique, we proposed and study a new accelerated algorithm for solving a convex bilevel optimization problem for which the inner level is the sum of smooth and nonsmooth convex functions and the outer level is a minimization of a smooth and strongly convex function over the set of solutions of the inner level. Then, we prove its strong convergence theorem under some conditions. As an application, we apply our proposed algorithm as a machine learning algorithm for solving some data classification problems. We also present some numerical experiments showing that our proposed algorithm has a better performance than the five other algorithms in the literature, namely BiG-SAM, iBiG-SAM, aiBiG-SAM, miBiG-SAM and amiBiG-SAM. [ABSTRACT FROM AUTHOR]

The path planning problem of nuclear environment robots refers to finding a collision-free path under the constraints of path length and an accumulated radiation dose. To solve this problem, the Improved Dueling Deep Double Q Network algorithm (ID3QN) based on asymmetric neural network structure was proposed. To address the issues of overestimation and low sample utilization in the traditional Deep Q Network (DQN) algorithm, we optimized the neural network structure and used the double network to estimate action values. We also improved the action selection mechanism, adopted a priority experience replay mechanism, and redesigned the reward function. To evaluate the efficiency of the proposed algorithm, we designed simple and complex radioactive grid environments for comparison. We compared the ID3QN algorithm with traditional algorithms and some deep reinforcement learning algorithms. The simulation results indicate that in the simple radioactive grid environment, the ID3QN algorithm outperforms traditional algorithms such as A*, GA, and ACO in terms of path length and accumulated radiation dosage. Compared to other deep reinforcement learning algorithms, including DQN and some improved DQN algorithms, the ID3QN algorithm reduced the path length by 15.6%, decreased the accumulated radiation dose by 23.5%, and converged approximately 2300 episodes faster. In the complex radioactive grid environment, the ID3QN algorithm also outperformed the A*, GA, ACO, and other deep reinforcement learning algorithms in terms of path length and an accumulated radiation dose. Furthermore, the ID3QN algorithm can plan an obstacle-free optimal path with a low radiation dose even in complex environments. These results demonstrate that the ID3QN algorithm is an effective approach for solving robot path planning problems in nuclear environments, thereby enhancing the safety and reliability of robots in such environments. [ABSTRACT FROM AUTHOR]

Data clustering is a complex data mining problem that clusters a massive amount of data objects into a predefined number of clusters; in other words, it finds symmetric and asymmetric objects. Various optimization methods have been used to solve different machine learning problems. They usually suffer from local optimal problems and unbalance between the search mechanisms. This paper proposes a novel hybrid optimization method for solving various optimization problems. The proposed method is called HRSA, which combines the original Reptile Search Algorithm (RSA) and Remora Optimization Algorithm (ROA) and handles these mechanisms' search processes by a novel transition method. The proposed HRSA method aims to avoid the main weaknesses raised by the original methods and find better solutions. The proposed HRSA is tested on solving various complicated optimization problems—twenty-three benchmark test functions and eight data clustering problems. The obtained results illustrate that the proposed HRSA method performs significantly better than the original and comparative state-of-the-art methods. The proposed method overwhelmed all the comparative methods according to the mathematical problems. It obtained promising results in solving the clustering problems. Thus, HRSA has a remarkable efficacy when employed for various clustering problems. [ABSTRACT FROM AUTHOR]

Many real-world scenarios can be naturally modeled as heterogeneous graphs, which contain both symmetry and asymmetry information. How to learn useful knowledge from the graph has become one of the hot spots of research in artificial intelligence. Based on Metapath2vec algorithm, an improved Metapath2vec algorithm is presented, which combines Metapath random walk, used to capture semantics and structure information between different nodes of a heterogeneous network, and GloVe model to consider the advantage of global text representation. In order to verify the feasibility and effectiveness of the model, node clustering and link prediction experiments were conducted on the self-generated ideal dataset and the MOOC course data. The analysis of experimental data on these tasks shows that the Metapath–GloVe algorithm learns consistently better embedding of heterogeneous nodes, and the algorithm improves the node embedding performance to better characterize the heterogeneous network structure and learn the characteristics of nodes, which proves the effectiveness and scalability of the proposed method in heterogeneous network mining tasks. It is also shown through extensive experiments that the Metapath–GloVe algorithm is more efficient than the non-negative matrix decomposition algorithm (NMF), and it can obtain better clustering results and more accurate prediction results in the video recommendation task. [ABSTRACT FROM AUTHOR]

In solving classification problems in the field of machine learning and pattern recognition, the pre-processing of data is particularly important. The processing of high-dimensional feature datasets increases the time and space complexity of computer processing and reduces the accuracy of classification models. Hence, the proposal of a good feature selection method is essential. This paper presents a new algorithm for solving feature selection, retaining the selection and mutation operators from traditional genetic algorithms. On the one hand, the global search capability of the algorithm is ensured by changing the population size, on the other hand, finding the optimal mutation probability for solving the feature selection problem based on different population sizes. During the iteration of the algorithm, the population size does not change, no matter how many transformations are made, and is the same as the initialized population size; this spatial invariance is physically defined as symmetry. The proposed method is compared with other algorithms and validated on different datasets. The experimental results show good performance of the algorithm, in addition to which we apply the algorithm to a practical Android software classification problem and the results also show the superiority of the algorithm. [ABSTRACT FROM AUTHOR]

Due to the insidious characteristics of network intrusion behaviors, developing an efficient intrusion detection system is still a big challenge, especially in the era of big data where the number of traffic and the dimension of each traffic feature are high. Because of the shortcomings of traditional common machine learning algorithms in network intrusion detection, such as insufficient accuracy, a network intrusion detection system based on LightGBM and autoencoder (AE) is proposed. The LightGBM-AE model proposed in this paper includes three steps: data preprocessing, feature selection, and classification. The LightGBM-AE model adopts the LightGBM algorithm for feature selection, and then uses an autoencoder for training and detection. When a set of data containing network intrusion behaviors are inputted into an autoencoder, there is a large reconstruction error between the original input data and the reconstructed data obtained by the autoencoder, which provides a basis for intrusion detection. According to the reconstruction error, an appropriate threshold is set to distinguish symmetrically between normal behavior and attack behavior. The experiment is carried out on the NSL-KDD dataset and implemented using Pytorch. In addition to autoencoder, variational autoencoder (VAE) and denoising autoencoder (DAE) are also used for intrusion detection and are compared with existing machine learning algorithms such as Decision Tree, Random Forest, KNN, GBDT, and XGBoost. The evaluation is carried out through classification evaluation indexes such as accuracy, precision, recall, F1-score. The experimental results show that the method can efficiently separate the attack behavior from normal behavior according to the reconstruction error. Compared with other methods, the effectiveness and superiority of this method are verified. [ABSTRACT FROM AUTHOR]

In this paper, a stochastic quasi-Newton algorithm for nonconvex stochastic optimization is presented. It is derived from a classical modified BFGS formula. The update formula can be extended to the framework of limited memory scheme. Numerical experiments on some problems in machine learning are given. The results show that the proposed algorithm has great prospects. [ABSTRACT FROM AUTHOR]

The effect of the application of machine learning on data streams is influenced by concept drift, drift deviation, and noise interference. This paper proposes a data stream anomaly detection algorithm combined with control chart and sliding window methods. This algorithm is named DCUSUM-DS (Double CUSUM Based on Data Stream), because it uses a dual mean value cumulative sum. The DCUSUM-DS algorithm based on nested sliding windows is proposed to satisfy the concept drift problem; it calculates the average value of the data within the window twice, extracts new features, and then calculates accumulated and controlled graphs to avoid misleading by interference points. The new algorithm is simulated using drilling engineering industrial data. Compared with automatic outlier detection for data streams (A-ODDS) and with sliding nest window chart anomaly detection based on data streams (SNWCAD-DS), the DCUSUM-DS can account for concept drift and shield a small amount of interference deviating from the overall data. Although the algorithm complexity increased from 0.1 second to 0.19 second, the classification accuracy receiver operating characteristic (ROC) increased from 0.89 to 0.95. This meets the needs of the oil drilling industry data stream with a sampling frequency of 1 Hz, and it improves the classification accuracy. [ABSTRACT FROM AUTHOR]

Since data stream anomaly detection algorithms based on sliding windows are sensitive to the abnormal deviation of individual interference data, this paper presents a sliding nest window chart anomaly detection based on the data stream (SNWCAD-DS) by employing the concept of the sliding window and control chart. By nesting a small sliding window in a large sliding window and analyzing the deviation distance between the small window and the large sliding window, the algorithm increases the out-of-bounds detection ratio and classifies the conceptual drift data stream online. The designed algorithm is simulated on the industrial data stream of drilling engineering. The proposed algorithm SNWCAD is compared with Automatic Outlier Detection for Data Streams (A-ODDS) and Distance-Based Outline Detection for Data Stream (DBOD-DS). The experimental results show that the new algorithm can obtain higher detection accuracy than the compared algorithms. Furthermore, it can shield the influence of individual interference data and satisfy actual engineering needs. [ABSTRACT FROM AUTHOR]

In order to protect people's lives and property, increasing numbers of explosive disposal robots have been developed. It is necessary for an explosive ordinance disposal (EOD) robot to quickly detect all explosives, especially when the location of the explosives is unknown. To achieve this goal, we propose a bidirectional dynamic weighted-A star (BD-A*) algorithm and a learn memory-swap sequence particle swarm optimization (LM-SSPSO) algorithm. Firstly, in the BD-A* algorithm, our aim is to obtain the shortest distance path between any two goal positions, considering computation time optimization. We optimize the computation time by introducing a bidirectional search and a dynamic OpenList cost weight strategy. Secondly, the search-adjacent nodes are extended to obtain a shorter path. Thirdly, by using the LM-SSPSO algorithm, we aim to plan the shortest distance path that traverses all goal positions. The problem is similar to the symmetric traveling salesman problem (TSP). We introduce the swap sequence strategy into the traditional PSO and optimize the whole PSO process by imitating human learning and memory behaviors. Fourthly, to verify the performance of the proposed algorithm, we begin by comparing the improved A* with traditional A* over different resolutions, weight coefficients, and nodes. The hybrid PSO algorithm is also compared with other intelligent algorithms. Finally, different environment maps are also discussed to further verify the performance of the algorithm. The simulation results demonstrate that our improved A* algorithm has superior performance by finding the shortest distance with less computational time. In the simulation results for LM-SSPSO, the convergence rate significantly improves, and the improved algorithm is more likely to obtain the optimal path. [ABSTRACT FROM AUTHOR]

Although linearly combining multiple variables can provide adequate diagnostic performance, certain algorithms have the limitation of being computationally demanding when the number of variables is sufficiently high. Liu et al. proposed the min–max approach that linearly combines the minimum and maximum values of biomarkers, which is computationally tractable and has been shown to be optimal in certain scenarios. We developed the Min–Max–Median/IQR algorithm under Youden index optimisation which, although more computationally intensive, is still approachable and includes more information. The aim of this work is to compare the performance of these algorithms with well-known Machine Learning algorithms, namely logistic regression and XGBoost, which have proven to be efficient in various fields of applications, particularly in the health sector. This comparison is performed on a wide range of different scenarios of simulated symmetric or asymmetric data, as well as on real clinical diagnosis data sets. The results provide useful information for binary classification problems of better algorithms in terms of performance depending on the scenario. [ABSTRACT FROM AUTHOR]

Traditional data analytics tools are designed to deal with the asymmetrical type of data i.e., structured, semi-structured, and unstructured. The diverse behavior of data produced by different sources requires the selection of suitable tools. The restriction of recourses to deal with a huge volume of data is a challenge for these tools, which affects the performances of the tool's execution time. Therefore, in the present paper, we proposed a time optimization model, shares common HDFS (Hadoop Distributed File System) between three Name-node (Master Node), three Data-node, and one Client-node. These nodes work under the DeMilitarized zone (DMZ) to maintain symmetry. Machine learning jobs are explored from an independent platform to realize this model. In the first node (Name-node 1), Mahout is installed with all machine learning libraries through the maven repositories. The second node (Name-node 2), R connected to Hadoop, is running through the shiny-server. Splunk is configured in the third node (Name-node 3) and is used to analyze the logs. Experiments are performed between the proposed and legacy model to evaluate the response time, execution time, and throughput. K-means clustering, Navies Bayes, and recommender algorithms are run on three different data sets, i.e., movie rating, newsgroup, and Spam SMS data set, representing structured, semi-structured, and unstructured data, respectively. The selection of tools defines data independence, e.g., Newsgroup data set to run on Mahout as others cannot be compatible with this data. It is evident from the outcome of the data that the performance of the proposed model establishes the hypothesis that our model overcomes the limitation of the resources of the legacy model. In addition, the proposed model can process any kind of algorithm on different sets of data, which resides in its native formats. [ABSTRACT FROM AUTHOR]

The existing learning-based unsupervised hashing method usually uses a pre-trained network to extract features, and then uses the extracted feature vectors to construct a similarity matrix which guides the generation of hash codes through gradient descent. Existing research shows that the algorithm based on gradient descent will cause the hash codes of the paired images to be updated toward each other's position during the training process. For unsupervised training, this situation will cause large fluctuations in the hash code during training and limit the learning efficiency of the hash code. In this paper, we propose a method named Deep Unsupervised Hashing with Gradient Attention (UHGA) to solve this problem. UHGA mainly includes the following contents: (1) use pre-trained network models to extract image features; (2) calculate the cosine distance of the corresponding features of the pair of images, and construct a similarity matrix through the cosine distance to guide the generation of hash codes; (3) a gradient attention mechanism is added during the training of the hash code to pay attention to the gradient. Experiments on two existing public datasets show that our proposed method can obtain more discriminating hash codes. [ABSTRACT FROM AUTHOR]

The concept of trend in data and a novel neural network method for the forecasting of upcoming time-series data are proposed in this paper. The proposed method extracts two data sets—the trend and the remainder—resulting in two separate learning sets for training. This method works sufficiently, even when only using a simple recurrent neural network (RNN). The proposed scheme is demonstrated to achieve better performance in selected real-life examples, compared to other averaging-based statistical forecast methods and other recurrent methods, such as long short-term memory (LSTM). [ABSTRACT FROM AUTHOR]

Since 2016, Mirai and Persirai malware have infected hundreds of thousands of Internet of Things (IoT) devices and created a massive IoT botnet, which caused distributed denial of service (DDoS) attacks. IoT malware targets vulnerable IoT devices, which are vulnerable to security risks. Techniques are needed to prevent IoT devices from being exploited by attackers. However, unlike high-performance PCs, IoT devices are lightweight, low-power, and low-cost, having performance limitations regarding processing and memory, which makes it difficult to install security and anti-malware programs. Recently, several studies have been attempted to quickly search for vulnerable internet-connected devices to solve this real issue. Issues yet to be studied still exist regarding these types of internet-wide scan technologies, such as filtering by security devices and a shortage of collected operating system (OS) information. This paper proposes an intelligent internet-wide scan model that improves IP state scanning with advanced internet protocol (IP) randomization, reactive protocol (port) scanning, and OS fingerprinting scanning, applying k* algorithm in order to find vulnerable IoT devices. Additionally, we describe the experiment’s results compared to the existing internet-wide scan technologies, such as ZMap and Shodan. As a result, the proposed model experimentally shows improved performance. Although we improved the ZMap, the throughput per minute (TPM) performance is similar to ZMap without degrading the IP scan throughput and the performance of generating a single IP address is about 118% better than ZMap. In the protocol scan performance experiments, it is about 129% better than the Censys based ZMap, and the performance of OS fingerprinting is better than ZMap, with about 50% accuracy. [ABSTRACT FROM AUTHOR]

The research on autonomous driving based on deep reinforcement learning algorithms is a research hotspot. Traditional autonomous driving requires human involvement, and the autonomous driving algorithms based on supervised learning must be trained in advance using human experience. To deal with autonomous driving problems, this paper proposes an improved end-to-end deep deterministic policy gradient (DDPG) algorithm based on the convolutional block attention mechanism, and it is called multi-input attention prioritized deep deterministic policy gradient algorithm (MAPDDPG). Both the actor network and the critic network of the model have the same structure with symmetry. Meanwhile, the attention mechanism is introduced to help the vehicles focus on useful environmental information. The experiments are conducted in the open racing car simulator (TORCS)and the results of five experiment runs on the test tracks are averaged to obtain the final result. Compared with the state-of-the-art algorithm, the maximum reward increases from 62,207 to 116,347, and the average speed increases from 135 km/h to 193 km/h, while the number of success episodes to complete a circle increases from 96 to 147. Also, the variance of the distance from the vehicle to the center of the road is compared, and the result indicates that the variance of the DDPG is 0.6 m while that of the MAPDDPG is only 0.2 m. The above results indicate that the proposed MAPDDPG achieves excellent performance. [ABSTRACT FROM AUTHOR]

Imbalanced data and feature redundancies are common problems in many fields, especially in software defect prediction, data mining, machine learning, and industrial big data application. To resolve these problems, we propose an intelligent fusion algorithm, SMPSO-HS-AdaBoost, which combines particle swarm optimization based on subgroup migration and adaptive boosting based on hybrid-sampling. In this paper, we apply the proposed intelligent fusion algorithm to software defect prediction to improve the prediction efficiency and accuracy by solving the issues caused by imbalanced data and feature redundancies. The results show that the proposed algorithm resolves the coexisting problems of imbalanced data and feature redundancies, and ensures the efficiency and accuracy of software defect prediction. [ABSTRACT FROM AUTHOR]

To provide an available diagnostic model for diagnosing coronary atherosclerotic heart disease to provide an auxiliary function for doctors, we proposed a new evolutionary classification model in this paper. The core of the prediction model is a kernel extreme learning machine (KELM) optimized by an improved salp swarm algorithm (SSA). To get a better subset of parameters and features, the space transformation mechanism is introduced in the optimization core to improve SSA for obtaining an optimal KELM model. The KELM model for the diagnosis of coronary atherosclerotic heart disease (STSSA-KELM) is developed based on the optimal parameters and a subset of features. In the experiment, STSSA-KELM is compared with some widely adopted machine learning methods (MLM) in coronary atherosclerotic heart disease prediction. The experimental results show that STSSA-KELM can realize excellent classification performance and more robust stability under four indications. We also compare the convergence of STSSA-KELM with other MLM; the STSSA-KELM model has demonstrated a higher classification performance. Therefore, the STSSA-KELM model can effectively help doctors to diagnose coronary heart disease. [ABSTRACT FROM AUTHOR]

Optical character recognition is gaining immense importance in the domain of deep learning. With each passing day, handwritten digits (0–9) data are increasing rapidly, and plenty of research has been conducted thus far. However, there is still a need to develop a robust model that can fetch useful information and investigate self-build handwritten digit data efficiently and effectively. The convolutional neural network (CNN) models incorporating a sigmoid activation function with a large number of derivatives have low efficiency in terms of feature extraction. Here, we designed a novel CNN model integrated with the extreme learning machine (ELM) algorithm. In this model, the sigmoid activation function is upgraded as the rectified linear unit (ReLU) activation function, and the CNN unit along with the ReLU activation function are used as a feature extractor. The ELM unit works as the image classifier, which makes the perfect symmetry for handwritten digit recognition. A deeplearning4j (DL4J) framework-based CNN-ELM model was developed and trained using the Modified National Institute of Standards and Technology (MNIST) database. Validation of the model was performed through self-build handwritten digits and USPS test datasets. Furthermore, we observed the variation of accuracies by adding various hidden layers in the architecture. Results reveal that the CNN-ELM-DL4J approach outperforms the conventional CNN models in terms of accuracy and computational time. [ABSTRACT FROM AUTHOR]

Narrow markets are typically considered those that due to limited liquidity or peculiarities in its investor base, such as a particularly high concentration of retail investors, make the stock market less efficient and arguably less predictable. We show in this article that neural networks, applied to narrow markets, can provide relatively accurate forecasts in narrow markets. However, practical considerations such as potentially suboptimal trading infrastructure and stale prices should be taken into considerations. There is ample existing literature describing the use of neural network as a forecasting tool in deep stock markets. The application of neural networks to narrow markets have received much less literature coverage. It is however an important topic as having reliable stock forecasting tools in narrow markets can help with the development of the local stock market, potentially also helping the real economy. Neural networks applied to moderately narrow markets generated forecasts that appear to be comparable, but typically not as accurate, as those obtained in deep markets. These results are consistent across a wide range of learning algorithms and other network features such as the number of neurons. Selecting the appropriate network structure, including deciding what training algorithm to use, is a crucial step in order to obtain accurate forecasts. [ABSTRACT FROM AUTHOR]

Bankruptcy prediction has been a popular and challenging research topic in both computer science and economics due to its importance to financial institutions, fund managers, lenders, governments, as well as economic stakeholders in recent years. In a bankruptcy dataset, the problem of class imbalance, in which the number of bankruptcy companies is smaller than the number of normal companies, leads to a standard classification algorithm that does not work well. Therefore, this study proposes a cluster-based boosting algorithm as well as a robust framework using the CBoost algorithm and Instance Hardness Threshold (RFCI) for effective bankruptcy prediction of a financial dataset. This framework first resamples the imbalance dataset by the undersampling method using Instance Hardness Threshold (IHT), which is used to remove the noise instances having large IHT value in the majority class. Then, this study proposes a Cluster-based Boosting algorithm, namely CBoost, for dealing with the class imbalance. In this algorithm, the majority class will be clustered into a number of clusters. The distance from each sample to its closest centroid will be used to initialize its weight. This algorithm will perform several iterations for finding weak classifiers and combining them to create a strong classifier. The resample set resulting from the previous module, will be used to train CBoost, which will be used to predict bankruptcy for the validation set. The proposed framework is verified by the Korean bankruptcy dataset (KBD), which has a very small balancing ratio in both the training and the testing phases. The experimental results of this research show that the proposed framework achieves 86.8% in AUC (area under the ROC curve) and outperforms several methods for dealing with the imbalanced data problem for bankruptcy prediction such as GMBoost algorithm, the oversampling-based method using SMOTEENN, and the clustering-based undersampling method for bankruptcy prediction in the experimental dataset. [ABSTRACT FROM AUTHOR]

We proposed a new efficient image denoising scheme, which leads to four important contributions. The first is to integrate both reconstruction and learning based approaches into a single model so that we are able to benefit advantages from both approaches simultaneously. The second is to handle both multiplicative and additive noise removal problems. The third is that the proposed approach introduces a sparse term to reduce non-Gaussian outliers from multiplicative noise and uses a Laplacian Schatten norm to capture the global structure information. In addition, the image is represented by preserving the intrinsic local similarity via a sparse coding method, which allows our model to incorporate both global and local information from the image. Finally, we propose a new method that combines Method of Optimal Directions (MOD) with Approximate K-SVD (AK-SVD) for dictionary learning. Extensive experimental results show that the proposed scheme is competitive against some of the state-of-the-art denoising algorithms. [ABSTRACT FROM AUTHOR]