Performance Prediction of Differential Fibers with a Bi-Directional Optimization Approach.

Due to the limited generality of traditional fault diagnosis expert system and its low accuracy of extracting failure symptoms, a general fault monitoring and diagnosis expert system has been built. For different devices, users can build fault trees in an interactive way and then the fault trees will be saved as expert knowledge. A variety of sensors are fixed to monitor the real-time condition of the device and intelligent algorithms such as wavelet transform and neural network are used to assist the extraction of failure symptoms. On the basis of integration of multi-sensor failure symptoms, the fault diagnosis is realized through forward and backward reasoning. The simulation diagnosis experiments of NC device have shown the effectiveness of the proposed method.

We observe that the human trajectory is not only forward predictable, but also backward predictable. Both forward and backward trajectories follow the same social norms and obey the same physical constraints with the only difference in their time directions. Based on this unique property, we develop a new approach, called reciprocal learning, for human trajectory prediction. Two networks, forward and backward prediction networks, are tightly coupled, satisfying the reciprocal constraint, which allows them to be jointly learned. Based on this constraint, we borrow the concept of adversarial attacks of deep neural networks, which iteratively modifies the input of the network to match the given or forced network output, and develop a new method for network prediction, called reciprocal attack for matched prediction. It further improves the prediction accuracy. Our experimental results on benchmark datasets demonstrate that our new method outperforms the state-of-the-art methods for human trajectory prediction.

Retrosynthetic analysis is a canonical technique for planning the synthesis route of organic molecules in drug discovery and development. In this technique, the screening of synthetic tree branches requires accurate forward reaction prediction, but existing software is far from completing this step independently. Previous studies attempted to apply a neural network to forward reaction prediction, but the accuracy was not satisfying. Through using the Edit Vector-based description and extended-connectivity fingerprints to transform the reaction into a vector, this study focuses on the update of the neural network to improve the template-based forward reaction prediction. Hard-threshold activation and the target propagation algorithm are implemented by introducing mixed convex-combinatorial optimization. Comparative tests were conducted to explore the optimal hyperparameter set. Using 15,000 experimental reaction data extracted from granted United States patents, the proposed hard-threshold neural network was systematically trained and tested. The results demonstrated that a higher prediction accuracy was obtained than that for the traditional neural network with backpropagation algorithm. Some successfully predicted reaction examples are also briefly illustrated.

The newscast system has shifted from conventional print to online media platforms in the current computing era. As a result, online media platforms enable us to absorb information more quickly and with fewer editorial constraints, and false information is disseminated at an extraordinary rate and on a massive scale. Many practical algorithms for identifying fake News have recently been created, which use unidirectional text sequence analysis. News and social context-level information were encoded using sequential neural networks. As a result, a bidirectional training strategy is capable of enhancing classification. This paper proposed Deep Fake BERT, a new model for identifying bogus News in online media. The model uses a BERT-based deep learning technique by integrating multiple simultaneous modules into a single-layer DCNN with various kernel filter sizes and strides. This combination can handle ambiguity, the most challenging aspect of natural language comprehension. This approach used classification methods such as Naive Bayes, Feed Forward Neural Networks, and LSTM, and prediction results were compared. Based on the comparison, the proposed model yields a classification accuracy is 99.25% to the existing methods.

In recent years, multi-objective evolutionary optimization algorithms have shown success in different areas of research. Due to their efficiency and power, many researchers have concentrated on adapting evolutionary algorithms to generate Pareto solutions. This paper proposes a new memetic adaptive multi-objective evolutionary algorithm that is based on a three-term backpropagation network (MAMOT). This algorithm is an automatic search method for optimizing the parameters and performance of neural networks, and it relies on the use of the adaptive non-dominated sorting genetic algorithm-II integrated with the backpropagation algorithm, being used as a local search method. The presented MAMOT employs a self-adaptive mechanism toward improving the performance of the proposed algorithm and a local optimizer improving all the individuals in a population in order to obtain better accuracy and connection weights. In addition, it selects an appropriate number of hidden nodes simultaneously. The proposed method was applied to 11 datasets representing pattern classification problems, including two-class, multi-class and complex data reflecting real problems. Experiments were performed, and the results indicated that the proposed method is viable in pattern classification tasks compared to a multi-objective genetic algorithm based on a three-term backpropagation network (MOGAT) and some of the methods mentioned in the literature. The statistical analysis results of the t test and Wilcoxon signed-ranks test also show that the performance of the proposed method is significantly better than MOGAT.

Automatically identifying a cyclist as they pass the starting line in a track cycling race is primarily restricted to physical detection systems assigned to individual competitors. These current solutions are expensive and require physical hardware to be attached to the cyclists during a race. This paper proposes a method to both detect when a track cyclist passes the lap line and identify the cyclist using a region convolutional neural network (R-CNN) and K-means clustering. This method aims to overcome several limitations of prior research by identifying multiple cyclists in a single frame, requiring only a single image to train per cyclist, and increasing identification accuracy by combining a neural network with K-means clustering. The proposed method uses the Mask R-CNN with COCO 2017 dataset and K-means clustering to train the detector with a single image of each cyclist. Cyclists are then able to be identified during a race. The method achieved an average identification accuracy of 95.1% with single image training. Multiple cyclists could be detected in an individual frame through the successful integration of a combined Mask R-CNN K-means detector. The primary limitation of the method is the reduction in identification accuracy for cyclists wearing similar coloured clothing. By using K-means clustering and RGB range thresholding, the cyclists may be mis-identified if the colours are too similar. Further research that would benefit this solution includes developing additional points-of-interest to identify cyclists and additional cameras with different angles for instances of total overlap between cyclists.

In this paper, we propose a reconstruction of logic-based approaches to abductive reasoning in terms of ampliative adaptive logics. A main advantage of this reconstruction is that the resulting logics have a proof theory. As abductive reasoning is non-monotonic, the latter is necessarily dynamic (conclusions derived at some stage may at a later stage be rejected). The proof theory warrants, however, that the conclusions derived at a given stage are justified in view of the insight in the premises at that stage. Thus, it even leads to justified conclusions for undecidable fragments. Another advantage of the proposed logics is that they are much closer to natural reasoning than the existing systems. Usually, abduction is viewed as a form of “backward reasoning”. The search procedure by which this is realized (for instance, some form of linear linear resolution) is very different from the search procedures of human reasoners. The proposed logics treat abduction as a form of “forward reasoning” (Modus Ponens in the “wrong direction”). As a result, abductive steps are very natural, and are moreover nicely integrated with deductive steps. We present two new adaptive logics for abduction, and illustrate both with some examples from the history of the sciences (the discovery of Uranus and of Neptune). We also present some alternative systems that are better suited for non-creative forms of abductive reasoning.

Unsafe behavior is a critical factor leading to construction accidents. Despite numerous studies supporting this viewpoint, the process by which accidents are influenced by construction workers’ unsafe behaviors and the extent to which unsafe behaviors are involved in this process remain poorly discussed. Therefore, this paper selects cases from Chinese building construction accidents to explore the probabilistic transmission paths from unsafe behaviors to accidents using a Bayesian network. First, a list of unsafe behaviors is constructed based on safety standards and operating procedures. Second, several chains of unsafe behaviors are extracted from 287 accident cases within four types (fall, collapse, struck-by and lifting) to form a Bayesian network model. Finally, two accidents are specifically analyzed to verify the rationality of the proposed model through forward reasoning. Additionally, critical groups of unsafe behaviors leading to the four types of accidents are identified through backward reasoning. The results show the following: (i) The time sequence of unsafe behaviors in a chain does not affect the final posterior probability of an accident, but the accident attribute strength of an unsafe behavior, affects the growth rate of the posterior probability of an accident. (ii) The four critical groups of unsafe behaviors leading to fall, collapse, struck-by, and lifting are identified. This study is of theoretical and practical significance for on-site behavioral management and accident prevention.

We observe that human poses exhibit strong group-wise structural correlation and spatial coupling between keypoints due to the biological constraints of different body parts. This group-wise structural correlation can be explored to improve the accuracy and robustness of human pose estimation. In this work, we develop a self-constrained prediction-verification network to characterize and learn the structural correlation between keypoints during training. During the inference stage, the feedback information from the verification network allows us to perform further optimization of pose prediction, which significantly improves the performance of human pose estimation. Specifically, we partition the keypoints into groups according to the biological structure of human body. Within each group, the keypoints are further partitioned into two subsets, high-accuracy proximal keypoints and low-accuracy distal keypoints. We develop a self-constrained prediction-verification network to perform forward and backward predictions between these keypoint subsets. One fundamental challenge in pose estimation, as well as in generic prediction tasks, is that there is no mechanism for us to verify if the obtained pose estimation or prediction results are accurate or not, since the ground truth is not available. Once successfully learned, the verification network serves as an accuracy verification module for the forward pose prediction. During the inference stage, it can be used to guide the local optimization of the pose estimation results of low-accuracy keypoints with the self-constrained loss on high-accuracy keypoints as the objective function. Our extensive experimental results on benchmark MS COCO and CrowdPose datasets demonstrate that the proposed method can significantly improve the pose estimation results.KeywordsHuman pose estimationSelf-constrainedStructural inferencePrediction optimization

The textile industry uses natural or synthetic fibers, or both, as raw materials. Global climate change affects the production potential of natural fibers by increasing pressure on air, soil, and water resources. Consequently, the production of these fibers, which the sector requires, faces significant sustainability challenges. This literature review comparatively examines the natural fibers commonly used in the textile sector cotton, jute, flax, and wool, based on environmental, economic, and social sustainability criteria. Each fiber's production process was evaluated, along with factors such as water consumption, chemical input requirements, carbon and water footprints, biodegradability, soil health impacts, and production conditions. The findings indicate that the sustainability performance of fibers varies significantly depending on fiber type, production method, and geographical conditions. This review aims to assess the impact of natural fibers on sustainable fashion and circular textile systems and to provide guidance for future research and industrial applications.

A fast sampling based evolutionary algorithm for million-dimensional multiobjective optimization

Internal leakage rate prediction and failure diagnosis of buried pipeline ball valve based on valve cavity pressure detection

Many-objective optimisation problems (MaOPs) cause serious difficulties for existing multi-objective evolutionary algorithms (MOEAs). One common way to alleviate these difficulties is to use objective dimensionality reduction. Most existing objective reduction methods are time-consuming because they require MOEAs to run numerous generations. Pareto corner search evolutionary algorithm (PCSEA) was proposed in [18] to speed up objective reduction methods by only seeking corner solutions instead of whole solutions. However, the PCSEA-based objective reduction method in [18] needs to predefine a threshold to select objectives which strongly depends on problems and is not straightforward to obtain. This paper proposes a new objective dimensionality reduction method by integrating PCSEA and principal component analysis (PCA). Thanks to combining advantages of PCSEA and PCA, the proposed method not only can be efficient to eliminate redundant objectives, but also not require to define any parameter in advanced. The experimental results also show that the proposed method can perform objective reduction more successfully than the PCSEA-based objective reduction method. The results further strengthen the links between evolutionary computation and machine learning to address optimization problems.

Most environmental selection strategies in multiobjective evolutionary algorithms (MOEAs) select solutions based on their objective function values. However, the objective evaluations of many real-world problems are very time-consuming. The use of a large number of objective evaluations will inevitably reduce the efficiency of MOEAs. This paper proposes a fuzzy classification-assisted environmental selection (FAES) scheme to reduce the number of objective evaluations of MOEAs. The proposed method uses a fuzzy classifier to choose promising solutions in environmental selection. In the proposed method, first, solutions in the previous generations are classified into two classes using the Pareto dominance relation. The non-dominated solutions are positive class, and the dominated solutions are negative class. Next, the classified solutions are used to build a fuzzy classifier. Then, the built classifier is used to predict the membership degree of each of the current and offspring solutions. Only the offspring solutions, whose membership degrees to the positive class are larger than their parents’, are evaluated. The offspring solutions with smaller membership degrees are discarded with no objective evaluations. Therefore, the number of objective evaluations can be reduced. Finally, the evaluated offspring solutions are used in the environmental selection together with the current solutions. The proposed FAES strategy is integrated into an MOEA in computational experiments. Experimental results show the efficiency of the proposed FAES on reducing the number of objective evaluations.

The paper describes a methodology for selecting relevant fuzzy implications in forward and backward reasoning. The proposed methodology is based on the functional representation of fuzzy implications and dependencies between fuzzy implications. This can be useful for the design of an inference engine based on the rule knowledge for a given rule-based system.