Article
Computer Science, Artificial Intelligence
Tullio Mancini, Hector Calvo-Pardo, Jose Olmo
Summary: This paper introduces a novel prediction model based on an ensemble of deep neural networks with extra randomness for improved accuracy and uncertainty computation. The method performs well in terms of mean square prediction error and accuracy of prediction intervals, outperforming state-of-the-art methods in experimental settings.
Article
Computer Science, Artificial Intelligence
Yuandu Lai, Yucheng Shi, Yahong Han, Yunfeng Shao, Meiyu Qi, Bingshuai Li
Summary: This paper introduces a new deep learning method that constructs prediction intervals for regression neural networks by simultaneously performing point estimation and uncertainty quantification. The method learns uncertainty without uncertainty labels through a novel loss function and considers epistemic uncertainty in an ensemble form. Experimental results show that the performance of the method is competitive with other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Haocheng Lei, Anthony Bellotti
Summary: By generating prediction intervals, the risk of wrong predictions in deep learning regression can be effectively controlled. Current methods can reduce the width of intervals, but cannot ensure capturing enough real labels. This study proposes a direct optimization method to improve the quality of prediction intervals.
Article
Computer Science, Artificial Intelligence
Antonio Alcantara, Ines M. Galvan, Ricardo Aler
Summary: As the relevance of probabilistic forecasting grows, the need for estimating multiple high-quality prediction intervals (PI) also increases. The Pareto Optimal Prediction Interval Hypernetwork (POPI-HN) approach treats the coverage-width trade-off as a multi-objective problem and obtains a complete set of Pareto Optimal solutions. POPI-HN can be trained through gradient descent without needing extra parameters and allows users to extract the PI with the required coverage.
APPLIED SOFT COMPUTING
(2023)
Article
Chemistry, Multidisciplinary
Hua Zhong, Li Xu
Summary: The paper introduces an all-batch loss function for constructing high-quality prediction intervals, which can be trained using the gradient descent method for various sample batch sizes. A high-quality prediction interval generation framework is proposed using dual feedforward neural networks structure, adaptable to regression analysis and other problems.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Jonas Busk, Peter Bjorn Jorgensen, Arghya Bhowmik, Mikkel N. Schmidt, Ole Winther, Tejs Vegge
Summary: In this study, a message passing neural network model is extended to incorporate both aleatoric and epistemic uncertainty in a unified framework, and the predictive distribution is recalibrated for improved accuracy. The proposed method is shown to accurately predict molecular properties with well calibrated uncertainty estimates in experimental settings.
MACHINE LEARNING-SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Jun Lu, Jinliang Ding, Changxin Liu, Tianyou Chai
Summary: This article proposes a Bayesian-learning-based sparse stochastic configuration network (BSSCN) to address the architecture complexity and ill-posed problems of neural networks when dealing with high-dimensional data. By using a Laplace distribution as the prior distribution of the output weights of SCN, a sparse approximate Gaussian posterior is obtained, facilitating the training process and providing analytical solutions for output weights. The proposed method also includes a hyperparameter estimation process and a bootstrap ensemble strategy for constructing prediction intervals (PIs), which have shown effectiveness in experiments.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Honglin Wen, Jie Gu, Jinghuan Ma, Lyuzerui Yuan, Zhijian Jin
Summary: A deep-learning forecasting model based on neural basis expansion analysis is proposed to improve short term load forecasting by narrowing the prediction interval. The model uses a doubly residual stacking strategy to decompose the forecasting task into sub-problems and applies conformal quantile regression for better theoretical coverage guarantee. Experiments show that the proposed model demonstrates improved performance in capturing load characteristics and providing narrow prediction intervals with nearly nominal coverage.
IEEE TRANSACTIONS ON SMART GRID
(2021)
Article
Operations Research & Management Science
Ziyi Huang, Henry Lam, Haofeng Zhang
Summary: Deep learning is used to generate high-quality prediction intervals (PIs) to quantify uncertainty in regression tasks, including simulation metamodeling. Most existing methods lack accurate information on conditional coverage, which may lead to unreliable predictions if it is significantly smaller than the marginal coverage. To address this, an end-to-end framework is proposed to output high-quality PIs and provide conditional coverage estimation. A new loss function is designed for implementation and theoretically justified using an exponential concentration bound. Evaluation results show competitive performance in terms of average PI width and accurate estimation of conditional coverage.
JOURNAL OF SYSTEMS SCIENCE AND SYSTEMS ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Oscar Cartagena, Francesco Trovo, Manuel Roveri, Doris Saez
Summary: This work aims to design a novel evolving fuzzy prediction interval for modeling nonlinear time-variant systems. It integrates a passive mechanism to update the model when new data are available and an active mechanism to trigger adaptation mechanisms when changes are detected in the data-generating process. The proposed solution is based on a prediction interval using fuzzy numbers to handle the uncertainty of a system and has been tested on synthetic and real data, confirming its effectiveness for modeling systems with dynamic changes over time.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yang Liu, Jun Zhao, Wei Wang, Witold Pedrycz
Summary: The article proposes an interval type-2 fuzzy granular neural network dynamic ensemble approach to address the challenge of providing reliable prediction intervals. This new method can automatically generate, prune, and merge neural networks, aiding in the real-time perception of nonstationary environments.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2021)
Article
Engineering, Civil
Kan Guo, Yongli Hu, Zhen Qian, Hao Liu, Ke Zhang, Yanfeng Sun, Junbin Gao, Baocai Yin
Summary: This paper introduces an optimized graph convolution recurrent neural network for traffic prediction, which can better explore the spatial and temporal information of traffic data and learns an optimized graph through a data-driven approach to reveal the latent relationship among road segments.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Biotechnology & Applied Microbiology
Elena-Niculina Dragoi, Kazem Godini, Ali Koolivand
Summary: This study modeled the composting process of oily sludge using neuro-evolutive methodology, successfully predicting the removal efficiency of TPH and OC. Experimental data validated the accuracy of the model, which can be used to reduce the cost of bioremediation.
ENVIRONMENTAL TECHNOLOGY & INNOVATION
(2021)
Article
Computer Science, Artificial Intelligence
Minjung Lee, Jinsoo Bae, Seoung Bum Kim
Summary: Data-driven soft sensors using deep learning models have shown superior predictive performance, but may face trustworthiness issues when dealing with unexpected situations or noisy input data. By introducing uncertainty-aware soft sensors based on Bayesian recurrent neural networks, the reliability of predictive uncertainty can be increased, allowing for interval prediction without compromising the predictive performance of the soft sensor.
ADVANCED ENGINEERING INFORMATICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Arpit Kapoor, Anshul Negi, Lucy Marshall, Rohitash Chandra
Summary: Cyclone track forecasting is a critical climate science problem, and machine learning methods, especially recurrent neural networks (RNNs), have shown promise in this field. However, these methods often lack uncertainty quantification. This paper proposes variational RNNs, which approximate the posterior distribution of parameters by minimizing the KL-divergence loss, for cyclone track and intensity prediction. The results demonstrate that variational RNNs provide a good approximation with uncertainty quantification while maintaining prediction accuracy.
ENVIRONMENTAL MODELLING & SOFTWARE
(2023)
Article
Engineering, Industrial
G. Sierra, M. Orchard, K. Goebel, C. Kulkarni
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2019)
Article
Engineering, Mechanical
Daniel Hulse, Christopher Hoyle, Kai Goebel, Irem Y. Tumer
JOURNAL OF MECHANICAL DESIGN
(2019)
Article
Automation & Control Systems
Oguz Bektas, Jeffrey A. Jones, Shankar Sankararaman, Indranil Roychoudhury, Kai Goebel
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2019)
Article
Engineering, Mechanical
Zhixiong Li, Kai Goebel, Dazhong Wu
JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER-TRANSACTIONS OF THE ASME
(2019)
Article
Public, Environmental & Occupational Health
Xiaoge Zhang, Sankaran Mahadevan, Kai Goebel
Article
Computer Science, Interdisciplinary Applications
Junchuan Shi, Tianyu Yu, Kai Goebel, Dazhong Wu
Summary: This study investigates the impact of diversity in base learners and extracted features on the performance of ensemble learning for predicting the remaining useful life of bearings. Experimental results show that selecting diverse features and base learners in different degradation stages significantly improves the performance of the proposed ensemble learning method.
JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING
(2021)
Article
Engineering, Mechanical
Marcia L. Baptista, Elsa M. P. Henriques, Kai Goebel
Summary: Detecting changes in system characteristics before failure is crucial for accurate remaining useful life estimation. Recurrent neural network techniques can be used to locate the elbow point of an asset, improving prognostic accuracy.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Engineering, Industrial
Gina Sierra, Elinirina Robinson, Kai Goebel
Summary: This paper investigates the use of prognostic information in decision-making processes, focusing on risk-informed thresholds for maintenance or operational setting changes. Sampling-based techniques are analyzed for their effectiveness in uncertainty propagation and analysis, with Latin Hypercube Sampling (LHS) showing no significant advantage over Monte Carlo Sampling (MCS) in terms of tail prediction with small sample sizes. A methodology combining MCS and Kernel Density Estimation (KDE) is explored for improving tail accuracy with reduced sample size in battery end-of-discharge data.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Computer Science, Artificial Intelligence
Marcia L. Baptista, Kai Goebel, Elsa M. P. Henriques
Summary: Maintenance decisions in domains like aeronautics rely heavily on predicting component and system failures using data-driven techniques. However, the lack of interpretability has been a challenge for these techniques. This study examines the correlation between features used in data-driven prognostic approaches and established metrics, using the SHAP model from explainable artificial intelligence. The results show that SHAP values closely align with the prognostic metrics, highlighting the significance of model complexity in explainability.
ARTIFICIAL INTELLIGENCE
(2022)
Article
Engineering, Industrial
Manuel Arias Chao, Chetan Kulkarni, Kai Goebel, Olga Fink
Summary: A novel hybrid framework is proposed to combine physics-based performance models with deep learning algorithms for prognostics of complex safety-critical systems, improving prediction horizon by 127% compared to purely data-driven approaches. Physics-based performance models are used to infer unobservable model parameters related to system health and combined with sensor readings as input to a deep neural network, demonstrating superior performance over traditional data-driven methods.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Computer Science, Information Systems
Madhav Mishra, Jesper Martinsson, Kai Goebel, Matti Rantatalo
Summary: The Bayesian hierarchical model (BHM) presented in the study allows for predicting bearing life by incorporating envelope acceleration data, degradation models, and prior knowledge of bearing rating life. The approach enables inference at different hierarchical levels and can be applied in various data scenarios. This method provides accurate estimates of bearing life at both individual and group levels, making it useful for predicting remaining useful life in different bearing conditions.
Article
Computer Science, Information Systems
Manuel Arias Chao, Chetan Kulkarni, Kai Goebel, Olga Fink
Summary: To facilitate the development of prognostics methods, a realistic dataset of aircraft engine run-to-failure trajectories under real flight conditions was developed, which can be utilized for both prognostics and fault diagnostics.
Article
Engineering, Multidisciplinary
Kai Goebel, Brian Smith, Anupa Bajwa
INTERNATIONAL JOURNAL OF PROGNOSTICS AND HEALTH MANAGEMENT
(2019)
Article
Multidisciplinary Sciences
Oguz Bektas, Jeffrey A. Jones, Shankar Sankararaman, Indranil Roychoudhury, Kai Goebel
Article
Engineering, Aerospace
Jueming Hu, Heinz Erzberger, Kai Goebel, Yongming Liu
JOURNAL OF AEROSPACE INFORMATION SYSTEMS
(2020)
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.