Article
Computer Science, Artificial Intelligence
Ping Yuan, Biao Wang, Zhizhong Mao
Summary: This study proposed a dynamic outlier ensemble method to relax the assumption of independent errors made by base detectors. Artificial outliers are generated using the concept of multiple classifier behavior to estimate competences, and validation sets are optimized to find more representative objects. Competences of base detectors are estimated using a probabilistic method, and a switching mechanism is proposed for robust detection results.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2021)
Article
Microscopy
Samuel Bitrus, Harald Fitzek, Eugen Rigger, Johannes Rattenberger, Doris Entner
Summary: This paper investigates the application of single classifiers and multiple classifier systems in correlative microscopy and demonstrates the feasibility and superiority of automated classification in this context.
Article
Energy & Fuels
Dongliang Gong, Ying Gao, Yalin Kou, Yurang Wang
Summary: This study aims to accurately predict the early cycle life of lithium-ion batteries using evolutionary computation techniques and machine learning approaches. The research results show that the fusion feature selection method performs the best in terms of cycle life early prediction performance.
JOURNAL OF ENERGY STORAGE
(2022)
Article
Computer Science, Artificial Intelligence
Sadegh Salesi, Georgina Cosma
Summary: Evolutionary Computation (EC) algorithms are powerful techniques for feature selection, but they often suffer from the stability issue of reaching different solutions in each run. This paper introduces a novel algorithm called Generalisation Power Analysis (GPA) to evaluate feature subsets based on their generalisation power over multiple classifiers, outperforming alternative methods in achieving high generalisation power. Despite requiring more computation time, using GPA during feature selection results in a robust prediction model developed with features not biased towards a specific classifier.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Jalal Sadoon Hameed Al-bayati, Burak Berk Ustundag
Summary: A novel approach based on grasshopper optimization algorithm and feature selection has been proposed for diagnosing plant leaf diseases. The algorithm can detect diseases at an early stage, thereby increasing crop survival and protection.
CMC-COMPUTERS MATERIALS & CONTINUA
(2021)
Review
Computer Science, Artificial Intelligence
Nan Li, Lianbo Ma, Tiejun Xing, Guo Yu, Chen Wang, Yingyou Wen, Shi Cheng, Shangce Gao
Summary: Machine learning (ML), the most promising paradigm for discovering deep knowledge from data, has been widely applied in practical applications such as recommender systems, virtual reality, and semantic segmentation. However, building high-quality ML systems for specific tasks is challenging due to the need for expert knowledge and high computation costs. This paper provides a comprehensive review of evolutionary machine learning (EML) methods, discussing concepts, taxonomy criteria, research problems, and limitations. The automatic design of ML using evolutionary computation is an increasingly popular research trend that can address the challenges of developing ML in large-scale practical applications.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Information Systems
Chao Lyu, Yuhui Shi, Lijun Sun
Summary: This paper empirically evaluates and compares the performance of common machine learning models in smoothing high-dimensional fitness landscapes. It proposes a data-driven multi-task optimization (DDMTO) framework to enhance the search abilities of evolutionary algorithms in complex solution spaces. Experimental results show that, by embedding an appropriate smoothing model into the DDMTO framework, the exploration ability and global optimization performance of evolutionary algorithms can be significantly improved without increasing the computational cost.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Yongna Yao
Summary: This article explores the application of machine learning algorithms in large-scale data, and optimizes the algorithm by combining Bayesian network structure learning and binary evolutionary algorithm, improving the classification effect and accuracy.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Weiyu Chen, Hisao Ishibuchi, Ke Shang
Summary: This article discusses the importance of subset selection in evolutionary multiobjective optimization and proposes efficient greedy algorithms based on submodular property. Computational experiments show that these algorithms are faster than the standard greedy algorithms and also contribute to the research on performance indicators.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)
Article
Environmental Sciences
Long Cui, Jiahua Zhang, Zhenjiang Wu, Lan Xun, Xiaopeng Wang, Shichao Zhang, Yun Bai, Sha Zhang, Shanshan Yang, Qi Liu
Summary: Wetlands in the Yellow River Delta are important and vulnerable due to tidal action and sediment deposits. A object-oriented approach with feature preference machine learning was used to classify the wetlands. A superpixel segmentation method using the watershed algorithm improved the classification accuracy. The random forest classifier combining superpixel segmentation and feature selection methods outperformed other pixel-based machine learning methods with a 91.74% overall accuracy and a kappa coefficient of 0.9078.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Hakan Ezgi Kiziloz
Summary: This study formally compares different classifier ensemble methods in the feature selection domain and finds that ensemble methods outperform single classifiers, albeit with longer execution time, and are more effective in minimizing the number of features.
Article
Computer Science, Artificial Intelligence
Wang Chao, Kai Wu, Jing Liu
Summary: Learning how to optimize AUC performance for imbalanced data has been a topic of interest. This paper proposes an evolutionary multitasking framework (EMTAUC) that utilizes information from different tasks to improve AUC performance.
IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE
(2022)
Article
Biology
Yazhou Ji, Beibei Shi, Yuanyuan Li
Summary: In this study, a machine learning framework called MSRUN-KELM was designed for the diagnosis of multiple myeloma (MM) using slime mould Runge Kutta Optimizer (MSRUN) and kernel extreme learning machine. The results showed that MSRUN and MSRUN-KELM performed well in MM diagnosis.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Abdullah Elen, Emre Avuclu
Summary: Machine learning algorithms play a crucial role in various fields, helping researchers and planners understand problems and improve strategies. The proposed method in this study, SVD, achieved higher classification accuracy compared to traditional and state-of-the-art methods by considering factors such as standard deviation and z-score.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Theory & Methods
Akbar Telikani, Amirhessam Tahmassebi, Wolfgang Banzhaf, Amir H. Gandomi
Summary: Evolutionary Computation approaches, inspired by nature, provide a reliable and effective way to address complex problems in real-world applications. They have been used to improve machine learning models and quality of results, contributing to addressing challenges in the field.
ACM COMPUTING SURVEYS
(2021)
Article
Chemistry, Analytical
Ibon Merino, Jon Azpiazu, Anthony Remazeilles, Basilio Sierra
Summary: In this study, transferring pretrained weights from 2D networks to their corresponding 3D versions improved the performance of 3D deep learning methods. EfficientNetB0 achieved the highest accuracy in industrial object recognition using extrusion, which is comparable to state-of-the art methods.
Article
Chemistry, Analytical
Arantzazu Florez, Elena Murga, Itziar Ortiz de Zarate, Arrate Jaureguibeitia, Arkaitz Artetxe, Basilio Sierra
Summary: The use of biosensors in the food industry is increasingly necessary for real-time measurement of different analytes in food. By modeling the kinetic reaction and adjusting an exponential decay model to the biosensor response, a novel mathematical approach is proposed to estimate the measurement output in advance, reducing the required measurement time by about 40% while maintaining low error rates to meet industry accuracy standards.
Article
Chemistry, Analytical
David Velasquez, Alejandro Sanchez, Sebastian Sarmiento, Camilo Velasquez, Mauricio Toro, Edwin Montoya, Helmuth Trefftz, Mikel Maiza, Basilio Sierra
Summary: Coffee Leaf Rust (CLR) is a fungal epidemic disease affecting coffee trees globally since the 1980s. An integrated cyber-physical data-collection system was developed using Remote Sensing and Wireless Sensor Networks to gather CLR data on a test bench coffee-crop. This system can automatically collect, structure, store, and transfer reliable multi-type data from various field sensors and cameras for CLR diagnosis.
Article
Computer Science, Information Systems
Jose Luis Outon, Ibon Merino, Ivan Villaverde, Aitor Ibarguren, Hector Herrero, Paul Daelman, Basilio Sierra
Summary: The SHERLOCK project aims to integrate an autonomous industrial mobile manipulator (AIMM) to perform cooperative tasks between a robot and a human. By combining autonomous navigation and 3D perception technology, AIMM achieved pose estimation and manipulation of industrial objects in a real environment with a success rate of 83.33%.
Article
Automation & Control Systems
Alberto Diez-Olivan, Patxi Ortego, Javier Del Ser, Itziar Landa-Torres, Diego Galar, David Camacho, Basilio Sierra
Summary: Industrial prognosis involves predicting failures of industrial assets based on data collected by IoT sensors, but concept drift can impact the data over time. To address this, contextual and operational changes must be detected and managed to trigger rapid model adaptation mechanisms. The proposed adaptive learning approach using a dendritic cell algorithm and neural network model demonstrated superior performance compared to other drift detectors and classification models in experimental results on a real-world industrial problem.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
David Montero, Naiara Aginako, Basilio Sierra, Marcos Nieto
Summary: The demand for real-time face clustering algorithms has increased in recent years, especially for security and surveillance purposes. However, current methods are not suitable for real-time applications and online methods are less accurate. To address these limitations, researchers propose an online gaussian mixture-based clustering method (OGMC) that reduces dependency on data order and size and can handle complex data distributions.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Review
Chemistry, Analytical
Jose Maria Martinez-Otzeta, Itsaso Rodriguez-Moreno, Inigo Mendialdua, Basilio Sierra
Summary: Random Sample Consensus (RANSAC) is a robust estimation method for models contaminated by outliers. It starts with sample selection, evaluates the adequacy of the estimation, and repeats the process until a stopping criterion is met. RANSAC is widely used in robotics, particularly for finding geometric shapes in point clouds or estimating camera view transformations.
Article
Computer Science, Artificial Intelligence
Arantzazu Florez, Itsaso Rodriguez-Moreno, Arkaitz Artetxe, Igor Garcia Olaizola, Basilio Sierra
Summary: Detecting changes in data streams is a crucial problem in Industry 4.0. Traditional machine learning algorithms are often static and lack the ability to generalize to new concepts, resulting in a deterioration of predictive performance when there is a change in data distribution. Drift detecting methods offer a solution to identify concept drift in data. This paper introduces CatSight, a new approach for detecting sudden or abrupt drift in industrial processes, which combines Common Spatial Patterns with traditional machine learning algorithms and demonstrates its effectiveness through evaluation on a real use case.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Engineering, Electrical & Electronic
Mikel Labayen, Xabier Mendialdua, Naiara Aginako, Basilio Sierra
Summary: The automation of railway operations is growing, but there are no accepted certification rules for computer vision and AI-enhanced perception technologies. To meet the needs for trusted AI solutions, a semi-automatic system based on virtual scenarios is being developed to evaluate performance under different visibility conditions.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Information Systems
Meritxell Gomez-Omella, Basilio Sierra, Susana Ferreiro
Summary: The Internet of Things (IoT) technologies play a crucial role in Industry 4.0, digitizing industries and services for enhanced productivity. This research proposes metrics to measure Data Quality (DQ) in streaming time series and implements techniques and tools for monitoring and improving information quality.
Article
Computer Science, Information Systems
David Velasquez, Enrique Perez, Xabier Oregui, Arkaitz Artetxe, Jorge Manteca, Jordi Escayola Mansilla, Mauricio Toro, Mikel Maiza, Basilio Sierra
Summary: This research proposes a hybrid machine learning model for real-time fault and anomaly detection in industrial systems. The results show that the model performs well in improving anomaly detection.
Article
Computer Science, Information Systems
Itsaso Rodriguez-Moreno, Jose Maria Martinez-Otzeta, Izaro Goienetxea, Igor Rodriguez, Basilio Sierra
Summary: This paper introduces an approach using CSP algorithm to classify videos based on skeleton joints signals, and ultimately achieve action recognition through image classification techniques. The results of the tests indicate promising outcomes on two data sets.
Article
Computer Science, Information Systems
Mikel Labayen, Ricardo Vea, Julian Florez, Naiara Aginako, Basilio Sierra
Summary: Identity verification and proctoring are key challenges in online learning, especially for certification. Various technologies with different levels of automation are available to address these issues.
Article
Computer Science, Information Systems
Imanol Iriarte, Inaki Iglesias, Joseba Lasa, Hodei Calvo-Soraluze, Basilio Sierra
Summary: This article discusses the benefits of introducing a simple passive mechanism to enable rotor tilting in VTOL multirotor vehicles. By comparing two different architectures of vehicles in a vehicle simulator, it is shown that introducing a passive rotor tilting mechanism can improve vehicle stability, reduce energy consumption, increase passenger comfort, and improve position tracking precision.
Article
Computer Science, Information Systems
Amaia Gil, Marco Quartulli, Igor G. Olaizola, Basilio Sierra
Summary: This paper proposes a method for adaptively selecting optimal data points for sensor time series on a window-by-window basis, focusing on quantifying the effect of applying data selection algorithms to time series windows. The approach is validated on synthetically generated time series and the entire UCR time series public data archive.
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.