Article
Multidisciplinary Sciences
Vitor Hugo Serravalle Reis Rodrigues, Paulo Roberto de Melo Barros Junior, Euler Bentes dos Santos Marinho, Jose Luis Lima de Jesus Silva
Summary: Developing accurate models for groundwater control is crucial for managing and planning water resources from aquifer reservoirs. The proposed Wavelet Gated Multiformer combines the strengths of a vanilla Transformer and a Wavelet Crossformer to improve the model's predictive capabilities by computing the relationships between time-series points and finding trending periodic patterns. This model outperforms other transformer-like models in terms of Mean Absolute Error reduction.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Zhangjing Yang, Wei-Wu Yan, Xiaolin Huang, Lin Mei
Summary: This paper proposes a novel adaptive temporal-frequency network (ATFN) for mid- and long-term time series forecasting. The model combines deep learning networks and frequency patterns to learn the trend feature and capture dynamic periodic patterns of time series data. The experimental results demonstrate that the ATFN has promising performance and strong adaptability for long-term time-series forecasting.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Rakshitha Godahewa, Kasun Bandara, Geoffrey Webb, Slawek Smyl, Christoph Bergmeir
Summary: Ensembling techniques are used to improve the performance of Global Forecasting Models (GFM) and univariate models in heterogeneous datasets. A new clustered ensembles methodology is proposed to train multiple GFMs on different clusters of series, achieving higher accuracy than baseline models.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Engineering, Civil
Julien Monteil, Anton Dekusar, Claudio Gambella, Yassine Lassoued, Martin Mevissen
Summary: This work investigates the use of deep learning models for long-term large-scale traffic prediction tasks, focusing on scalability. By analyzing 14 weeks of speed observations from over 1000 segments in downtown Los Angeles, different machine learning and deep learning predictors were studied, along with their scalability to larger areas. The study shows that modeling temporal and spatial features into deep learning predictors can be beneficial for long-term predictions, while simpler predictors achieve satisfactory performance for link-based and short-term forecasting, with a trade-off in prediction accuracy, horizon, training time, and model sizing discussed.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Konstantinos Benidis, Syama Sundar Rangapuram, Valentin Flunkert, Yuyang Wang, Danielle Maddix, Caner Turkmen, Jan Gasthaus, Michael Bohlke-Schneider, David Salinas, Lorenzo Stella, Francois-Xavier Aubet, Laurent Callot, Tim Januschowski
Summary: Deep learning based forecasting methods have achieved remarkable success in time series prediction and have become widely used in industrial applications and forecasting competitions. This article provides an introduction to deep forecasting, discussing important building blocks and summarizing recent literature.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Artificial Intelligence
Ming Jin, Yu Zheng, Yuan-Fang Li, Siheng Chen, Bin Yang, Shirui Pan
Summary: This paper proposes a continuous model, MTGODE, to forecast multivariate time series by overcoming the limitations of discrete neural architectures, high complexity, and reliance on graph priors. MTGODE utilizes dynamic graph neural ordinary differential equations to unify spatial and temporal message passing, resulting in superior forecasting performance on benchmark datasets.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
ZhuoLin Li, Jie Yu, GaoWei Zhang, LingYu Xu
Summary: This paper proposes a novel dynamic spatio-temporal graph neural network (DSTGN) to tackle the challenge of predicting multivariate time series. The key components of DSTGN are dynamic graph estimation and adaptive guided propagation. Experimental results demonstrate that our method outperforms state-of-the-art baseline methods on four datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Engineering, Chemical
Zhu Wang, Laize Liu, Xiujuan Dong, Jiaxuan Liu
Summary: This paper proposes an integrated framework of neural network modelling and evaluation for nonlinear dynamic processes. The framework can handle noisy sensors and dense data, and employs two novel evaluation metrics to evaluate the model. Numerical experiments demonstrate the accuracy and stability of the framework.
CANADIAN JOURNAL OF CHEMICAL ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Philippe Chatigny, Jean-Marc Patenaude, Shengrui Wang
Summary: This study investigates the use of deep neural network models for forecasting in settings with complex time series behaviors. By building a multivariate autoregressive model and proposing a novel variable-length attention mechanism, the limitations of recurrent neural networks are effectively addressed. Experimental results demonstrate that the proposed approach significantly outperforms typical models in financial time series forecasting.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2021)
Article
Computer Science, Artificial Intelligence
Arun M. George, Sounak Dey, Dighanchal Banerjee, Arijit Mukherjee, Manan Suri
Summary: IoT-based automated systems require efficient online time-series analysis and forecasting, which is challenging to achieve on low-cost constrained edge devices. This study proposes a novel spiking reservoir based network that relies on temporal spike encoding and feedback-driven online learning mechanism for online time series forecasting. The network outperforms conventional methods like SARIMA, Online ARIMA, Stacked LSTM, achieving up to 8% higher R2 score while using negligible buffer memory.
Article
Computer Science, Artificial Intelligence
Artemios-Anargyros Semenoglou, Evangelos Spiliotis, Vassilios Assimakopoulos
Summary: Data augmentation techniques can improve forecasting accuracy in univariate time series prediction, especially when deep neural networks are used. However, these improvements become less significant as the initial size of the training set increases.
PATTERN RECOGNITION
(2022)
Article
Physics, Multidisciplinary
Kady Sako, Berthine Nyunga Mpinda, Paulo Canas Rodrigues
Summary: This study forecasts stock market indexes and currency exchange rates using Recurrent Neural Networks (RNNs) and its variants, with the Gated Recurrent Unit (GRU) model performing the best overall.
Article
Multidisciplinary Sciences
Yuzhen Zhu, Shaojie Luo, Di Huang, Weiyan Zheng, Fang Su, Beiping Hou
Summary: Recent studies have shown great performance of Transformer-based models in long-term time series forecasting, but they have limitations when training on small datasets. This paper proposes the DRCNN method to utilize the continuity between data by decomposing data into residual and trend terms, and designs the DR-Block to extract features. Additionally, a Multi-head Sequence method is proposed for longer inputs and accurate forecasts.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Zimeng Lyu, Alexander Ororbia, Travis Desell
Summary: Time series forecasting is an important task in data science, but offline-trained models often face data drift issues. To address this, this paper proposes an online neural architecture search algorithm (ONE-NAS) that can automatically design and train recurrent neural networks for online forecasting tasks. Experimental results show that ONE-NAS outperforms traditional statistical methods and using multiple populations of RNNs can significantly improve performance.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
D. Criado-Ramon, L. G. B. Ruiz, M. C. Pegalajar
Summary: This paper addresses the problem of electric demand prediction using neural networks and symbolization techniques. Symbolization techniques provide a shorter symbolic representation of a time series compared to the original time series. In the experimentation, the symbolization methodology resulted in a model that was trained significantly faster but had slightly worse quality metrics compared to the best numerical model.
APPLIED SOFT COMPUTING
(2022)
Article
Materials Science, Textiles
Jie Zhou, Xingxing Zou, Wai Keung Wong
Summary: This study addresses the challenges of automatic color sorting for waste textile recycling and introduces a computer vision-based color sorting system that efficiently classifies colors and meets the requirements of an automatic recycling line. The system shows good performance in classifying waste textile samples, providing valuable insights for improving the intelligent level of color sorting in textile recycling.
INTERNATIONAL JOURNAL OF CLOTHING SCIENCE AND TECHNOLOGY
(2022)
Article
Automation & Control Systems
Zhaolin Lai, Xiang Feng, Huiqun Yu, Fei Luo
Summary: Social spider optimization is an effective swarm algorithm for solving complex optimization problems by simulating cooperative behavior of spiders, but it suffers from long computation time and premature convergence on some problems. To improve search performance, a parallel SSO algorithm with emotional learning is proposed, which accelerates computation speed by parallel position update and enhances search ability by increasing swarm diversity.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Automation & Control Systems
Dongmei Mo, Wai Keung Wong, Zhihui Lai, Jie Zhou
Summary: This article proposes a weighted double-low-rank decomposition method to treat matrix singular values differently and preserve the most important characteristics of a fabric image for defect detection. This method is more robust and outperforms existing low-rank-based methods in locating defective regions on fabric images.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Dongmei Mo, Wai Keung Wong, Xianjing Liu, Yao Ge
Summary: In this paper, a concentrated hashing method with neighborhood embedding (CHNE) is proposed for efficient and effective image retrieval and classification. By integrating Cauchy cross-entropy and pair-wise weighted similarity loss, CHNE can improve the performance of retrieval and classification. The proposed method also addresses the problem of using binary codes for classification and enhances classification accuracy by minimizing the errors in the loss function.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2022)
Article
Automation & Control Systems
Jianjun Qian, Wai Keung Wong, Hengmin Zhang, Jin Xie, Jian Yang
Summary: This paper proposes a novel robust regression scheme by integrating Optimal Transport (OT) with convex regularization to handle structure noises in high-dimensional visual data. Experimental results demonstrate the superior performance of our method in biometric image classification compared to several state-of-the-art regression-based classification methods.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Artificial Intelligence
Jiajun Wen, Wai Keung Wong, Xiao-Li Hu, Honglin Chu, Zhihui Lai
Summary: This paper proposes a novel method called restricted subgradient descend for learning sparse signals. The method achieves accurate learning of high quality sparse solutions in finite iteration steps.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2022)
Article
Computer Science, Artificial Intelligence
Dongmei Mo, Xingxing Zou, WaiKeung Wong
Summary: Online stylist service has huge economic potentials due to the digitalization trend in the fashion industry. This paper proposes a visual and semantic representation model for explainable evaluation and recommendation, considering factors such as color, material, and style. The model can classify outfits into three precise evaluation levels and diagnose problematic fashion items.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Automation & Control Systems
Lunke Fei, Wai Keung Wong, Shuping Zhao, Jie Wen, Jian Zhu, Yong Xu
Summary: Palmprint recognition has garnered significant research interest due to its excellent contactless property and user-security, but existing methods are limited in practical application due to their focus on intraspectral palmprint recognition. This study proposes a spectrum-invariant feature learning method that addresses the problem of different spectra in gallery and probe samples. The method forms blockwise direction-based ordinal measure vectors and employs a unified feature projection to map different spectra of palmprint images into a common feature space, enhancing discriminative power while maintaining similarity in intraclass features. Experimental results demonstrate the method's effectiveness in cross-spectral palmprint recognition.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Computer Science, Information Systems
Jianjun Qian, Shumin Zhu, Chaoyu Zhao, Jian Yang, Wai Keung Wong
Summary: This paper proposes a hard samples guided optimal transport (OT) loss, OTFace, to improve face representation in the wild. It enhances the performance of hard samples by introducing feature distribution discrepancy while maintaining the performance on easy samples.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Yuwu Lu, Wai Keung Wong, Biqing Zeng, Zhihui Lai, Xuelong Li
Summary: This paper proposes a guided discrimination and correlation subspace learning (GDCSL) method for cross-domain image classification. GDCSL considers the domain-invariant, category-discriminative, and correlation learning of data. It introduces the discriminative information associated with the source and target data and extracts the most correlated features for image classification. Experimental results show the effectiveness of the proposed methods over state-of-the-art domain adaptation methods.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Computer Science, Information Systems
Yujuan Ding, Yunshan Ma, Lizi Liao, Wai Keung Wong, Tat-Seng Chua
Summary: Fashion trend forecasting is important for fashion companies and lovers. Previous studies focused on limited fashion elements and used statistical-based solutions, while this study proposes a neural network-based model called REAR, which considers the relations among fashion elements and user groups. Experimental results demonstrate the effectiveness of the REAR model in fashion trend forecasting.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Information Systems
Yujuan Ding, Yunshan Ma, Wai Keung Wong, Tat-Seng Chua
Summary: In this paper, a novel Attentional Content-level Translation-based Recommender (ACTR) framework is proposed to model the instant user intent and intent-specific transition probability for sequential fashion recommendation. The proposed method enhances the connectivity of fashion items and boosts the recommendation performance.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Yujuan Ding, Yunshan Ma, Wai Keung Wong, Tat-Seng Chua
Summary: Sequential fashion recommendation plays a significant role in online fashion shopping, with the key to building an effective model lying in capturing user preferences and item relationships. By leveraging global graphs and graph kernels for information propagation, user and item representations can be enhanced to improve the effectiveness and efficiency of sequential fashion recommendation.
PROCEEDINGS OF THE 2021 INTERNATIONAL CONFERENCE ON MULTIMEDIA RETRIEVAL (ICMR '21)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Pang Kaicheng, Zou Xingxing, Wai Keung Wong
Summary: The paper aims to model the fashion compatibility of an outfit and provide explanations using convolutional neural networks and Bi-LSTM model. Experimental results demonstrate the success of the proposed approach in evaluating compatibility and reasons behind it.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS, CVPRW 2021
(2021)
Article
Materials Science, Textiles
Xingxing Zou, Wai Keung Wong, Jianjun Qian
Summary: This study introduces a multi-domain fashion image recognition approach, which improves the flexibility of fashion image retrieval by establishing the Fashion-DA dataset and using an unsupervised domain adaption method based on adaptive feature norm. The effectiveness of the proposed method is evaluated through experiments.
AATCC JOURNAL OF RESEARCH
(2021)
Review
Management
Vinicius N. Motta, Miguel F. Anjos, Michel Gendreau
Summary: This survey presents a review of optimization approaches for the integration of demand response in power systems planning and highlights important future research directions.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Philipp Schulze, Armin Scholl, Rico Walter
Summary: This paper proposes an improved branch-and-bound algorithm, R-SALSA, for solving the simple assembly line balancing problem, which performs well in balancing workloads and providing initial solutions.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Roshan Mahes, Michel Mandjes, Marko Boon, Peter Taylor
Summary: This paper discusses appointment scheduling and presents a phase-type-based approach to handle variations in service times. Numerical experiments with dynamic scheduling demonstrate the benefits of rescheduling.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Oleg S. Pianykh, Sebastian Perez, Chengzhao Richard Zhang
Summary: Efficient scheduling is crucial for optimizing resource allocation and system performance. This study focuses on critical utilization and efficient scheduling in discrete scheduling systems, and compares the results with classical queueing theory.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Review
Management
Hamed Jahani, Babak Abbasi, Jiuh-Biing Sheu, Walid Klibi
Summary: Supply chain network design is a large and growing area of research. This study comprehensively surveys and analyzes articles published from 2008 to 2021 to detect and report financial perspectives in SCND models. The study also identifies research gaps and offers future research directions.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Patrick Healy, Nicolas Jozefowiez, Pierre Laroche, Franc Marchetti, Sebastien Martin, Zsuzsanna Roka
Summary: The Connected Max-k-Cut Problem is an extension of the well-known Max-Cut Problem, where the objective is to partition a graph into k connected subgraphs by maximizing the cost of inter-partition edges. The researchers propose a new integer linear program and a branch-and-cut algorithm for this problem, and also use graph isomorphism to structure the instances and facilitate their resolution. Extensive computational experiments show that, if k > 2, their approach outperforms existing algorithms in terms of quality.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Victor J. Espana, Juan Aparicio, Xavier Barber, Miriam Esteve
Summary: This paper introduces a new methodology based on the machine learning technique MARS for estimating production functions that satisfy classical production theory axioms. The new approach overcomes the overfitting problem of DEA through generalized cross-validation and demonstrates better performance in reducing mean squared error and bias compared to DEA and C2NLS methods.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Stefano Nasini, Rabia Nessah
Summary: In this paper, the authors investigate the impact of time flexibility in job scheduling, showing that it can significantly affect operators' ability to solve the problem efficiently. They propose a new methodology based on convex quadratic programming approaches that allows for optimal solutions in large-scale instances.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Zhiqiang Liao, Sheng Dai, Timo Kuosmanen
Summary: Nonparametric regression subject to convexity or concavity constraints is gaining popularity in various fields. The conventional convex regression method often suffers from overfitting and outliers. This paper proposes the convex support vector regression method to address these issues and demonstrates its advantages in prediction accuracy and robustness through numerical experiments.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Kuo-Hao Chang, Ying-Zheng Wu, Wen-Ray Su, Lee-Yaw Lin
Summary: The damage and destruction caused by earthquakes necessitates the evacuation of affected populations. Simulation models, such as the Stochastic Pedestrian Cell Transmission Model (SPCTM), can be utilized to enhance disaster and evacuation management. The analysis of SPCTM provides insights for government officials to formulate effective evacuation strategies.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Qinghua Wu, Mu He, Jin-Kao Hao, Yongliang Lu
Summary: This paper studies a variant of the orienteering problem known as the clustered orienteering problem. In this problem, customers are grouped into clusters and a profit is associated with each cluster, collected only when all customers in the cluster are served. The proposed evolutionary algorithm, incorporating a backbone-based crossover operator and a destroy-and-repair mutation operator, outperforms existing algorithms on benchmark instances and sets new records on some instances. It also demonstrates scalability on large instances and has shown superiority over three state-of-the-art COP algorithms. The algorithm is also successfully applied to a dynamic version of the COP considering stochastic travel time.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Bjorn Bokelmann, Stefan Lessmann
Summary: Estimating treatment effects is an important task for data analysts, and uplift models provide support for efficient allocation of treatments. However, evaluating uplift models is challenging due to variance issues. This paper theoretically analyzes the variance of uplift evaluation metrics, proposes variance reduction methods based on statistical adjustment, and demonstrates their benefits on simulated and real-world data.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Congzheng Liu, Wenqi Zhu
Summary: This paper proposes a feature-based non-parametric approach to minimizing the conditional value-at-risk in the newsvendor problem. The method is able to handle both linear and nonlinear profits without prior knowledge of the demand distribution. Results from numerical and real-life experiments demonstrate the robustness and effectiveness of the approach.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Laszlo Csato
Summary: This paper compares the performance of the eigenvalue method and the row geometric mean as two weighting procedures. Through numerical experiments, it is found that the priorities derived from the two eigenvectors in the eigenvalue method do not always agree, while the row geometric mean serves as a compromise between them.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
Article
Management
Guowei Dou, Tsan-Ming Choi
Summary: This study investigates the impact of channel relationships between manufacturers on government policies and explores the effectiveness of positive incentives versus taxes in increasing social welfare. The findings suggest that competition may be more effective in improving sustainability and social welfare. Additionally, government incentives for green technology may not necessarily enhance sustainability.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2024)
