4.7 Article

On-line transfer learning for multi-fidelity data fusion with ensemble of deep neural networks

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Computer Science, Artificial Intelligence

Counterfactual inference to predict causal knowledge graph for relational transfer learning by assimilating expert knowledge --Relational feature transfer learning algorithm

Jiarui Li et al.

Summary: Transfer learning is a machine learning method that transfers knowledge from existing models to solve new problems. This paper introduces a relational feature transfer learning algorithm that uses knowledge graphs and causal analysis to improve performance. Experimental results on healthcare-related datasets show that the algorithm can generate more accurate prediction models.

ADVANCED ENGINEERING INFORMATICS (2022)

Article Engineering, Multidisciplinary

Multi-fidelity regression using artificial neural networks: Efficient approximation of parameter-dependent output quantities

Mengwu Guo et al.

Summary: Multi-fidelity methods incorporate information from other sources to improve model results when time or budget constraints limit the generation of additional data.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2022)

Article Computer Science, Artificial Intelligence

How human-robot collaboration impacts construction productivity: An agent-based multi-fidelity modeling approach

Minghui Wu et al.

Summary: This research introduces an agent-based multi-fidelity modeling approach to simulate and evaluate the impact of human-robot collaboration (HRC) on construction productivity. The study finds that the supplement strategy and proactive interaction between robots and workers significantly improve productivity, and HRC has a scale effect on productivity per robot.

ADVANCED ENGINEERING INFORMATICS (2022)

Article Computer Science, Artificial Intelligence

An ensemble learning method based on deep neural network and group decision making

Xiaojun Zhou et al.

Summary: This paper proposes an ensemble learning method based on deep neural network and group decision making. Multiple deep neural networks are used to generate individual learners, and group decision making is applied to find the optimal alternative. Experimental results demonstrate the effectiveness and superiority of this method in image classification.

KNOWLEDGE-BASED SYSTEMS (2022)

Article Computer Science, Interdisciplinary Applications

Transfer learning based multi-fidelity physics informed deep neural network

Souvik Chakraborty

Summary: This paper introduces a novel multi-fidelity physics informed deep neural network (MF-PIDNN) to address the challenge of analyses and design for systems where the governing differential equation is either not known or known in an approximate sense. MF-PIDNN blends physics informed and data-driven deep learning techniques by using transfer learning, providing accurate predictions even in zones with no data. The approach updates a low-fidelity model with available high-fidelity data, showcasing the effectiveness of transfer learning in solving multi-fidelity problems.

JOURNAL OF COMPUTATIONAL PHYSICS (2021)

Article Computer Science, Artificial Intelligence

A surrogate-ensemble assisted expensive many-objective optimization

Yi Zhao et al.

Summary: This paper proposes a method to train multiple surrogate models to assist many-objective optimization algorithm for solving expensive many-objective problems. Experimental results show that this method is competitive with other surrogate-assisted evolutionary algorithms within a limited computational budget.

KNOWLEDGE-BASED SYSTEMS (2021)

Article Computer Science, Artificial Intelligence

A screening-based gradient-enhanced Gaussian process regression model for multi-fidelity data fusion

Quan Lin et al.

Summary: The proposed gradient-enhanced multi-fidelity Gaussian process model effectively reduces modeling complexity by selecting a portion of gradients from the low-fidelity model. Testing on multiple numerical examples and an aerodynamic airfoil case shows that the method significantly improves modeling efficiency with little loss in prediction accuracy.

ADVANCED ENGINEERING INFORMATICS (2021)

Article Engineering, Aerospace

A competitive variable-fidelity surrogate-assisted CMA-ES algorithm using data mining techniques

Zengcong Li et al.

Summary: The paper proposes a new CVFS-CMA-ES algorithm that utilizes data mining techniques to enhance its computational efficiency and global optimizing ability. By modifying the LCB method and introducing the step size control, the algorithm achieves a balance between exploration and exploitation, leading to higher efficiency and global optimizing ability compared to other algorithms.

AEROSPACE SCIENCE AND TECHNOLOGY (2021)

Article Mechanics

Machine learning regression approaches for predicting the ultimate buckling load of variable-stiffness composite cylinders

A. Kaveh et al.

Summary: In this study, various machine learning techniques were employed to establish the relationship between fiber angle and buckling capacity of cylinders under bending-induced loads. The numerical results demonstrate that compared with other models such as Random Forest Regressor, Decision Tree algorithm, and Multiple Linear Regression, the Deep Learning model achieved more stable results with smaller errors and higher generalization.

ACTA MECHANICA (2021)

Article Engineering, Multidisciplinary

Multi-fidelity deep neural network surrogate model for aerodynamic shape optimization

Xinshuai Zhang et al.

Summary: The study establishes an effective optimization framework of aerodynamic shape design based on the multi-fidelity deep neural network (MFDNN) model, achieving high accuracy by blending different fidelity information. The proposed framework significantly improves optimization efficiency and outperforms single-fidelity methods.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2021)

Article Mechanics

Transfer learning based variable-fidelity surrogate model for shell buckling prediction

Kuo Tian et al.

Summary: This paper proposes a novel establishment method of the variable-fidelity surrogate model driven by transfer learning for shell buckling prediction problems, aiming to achieve higher prediction accuracy. The TL-VFSM is constructed through two steps and verified by engineering examples for its effectiveness and efficiency.

COMPOSITE STRUCTURES (2021)

Article Mathematics, Applied

CONVERGENCE AND DYNAMICAL BEHAVIOR OF THE ADAM ALGORITHM FOR NONCONVEX STOCHASTIC OPTIMIZATION

Anas Barakat et al.

Summary: The paper introduces a continuous-time version and a version with decreasing step size of the Adam algorithm, proving their convergence under different conditions, analyzing their convergence rates, and fluctuations.

SIAM JOURNAL ON OPTIMIZATION (2021)

Article Engineering, Multidisciplinary

The role of surrogate models in the development of digital twins of dynamic systems

S. Chakraborty et al.

Summary: Digital twin technology shows great potential in various industrial sectors, but its adoption has been slow due to lack of clarity for specific applications. This paper explores the use of surrogate models within digital twin technology, particularly focusing on Gaussian process emulator. Numerical simulation results demonstrate the potential of surrogate models, such as GP emulators, to be effective tools for digital twin development.

APPLIED MATHEMATICAL MODELLING (2021)

Article Engineering, Mechanical

A Sequential Sampling Generation Method for Multi-Fidelity Model Based on Voronoi Region and Sample Density

Yin Liu et al.

Summary: The study introduces a sequential sampling generation method based on the Voronoi region and sample density (SSGM-VRDS) to gradually extend data sets and improve the accuracy of initial models by selecting new infilling samples during each iteration. The proposed method demonstrates outstanding effectiveness and efficiency in improving the accuracy of a surrogate model, as well as practicality for solving practical optimization problems.

JOURNAL OF MECHANICAL DESIGN (2021)

Article Engineering, Civil

Accelerated optimization of curvilinearly stiffened panels using deep learning

Karanpreet Singh et al.

Summary: The aerospace industry aims to design robust and fuel efficient structures, utilizing advanced manufacturing techniques like additive manufacturing. Approximating FEA buckling response can accelerate the optimization process, with trained DNN achieving 95% accuracy on the test set.

THIN-WALLED STRUCTURES (2021)

Article Computer Science, Artificial Intelligence

Dynamic parameter inverse analysis of concrete dams based on Jaya algorithm with Gaussian processes surrogate model

Fei Kang et al.

Summary: This study presents a dynamic parameter inverse analysis process for concrete dams based on Gaussian process regression and Jaya algorithm, and verifies the performance of this method through numerical examples. The results show that Gaussian process regression can significantly improve parameter identification efficiency, and the Jaya algorithm shows promise in parameter recognition.

ADVANCED ENGINEERING INFORMATICS (2021)

Article Engineering, Aerospace

Rapid airfoil design optimization via neural networks-based parameterization and surrogate modeling

Xiaosong Du et al.

Summary: The aerodynamic optimization based on CFD reduces design time significantly compared with manual design but can still take hours to converge. A fast, interactive design framework is proposed to complete airfoil aerodynamic optimization within seconds, using a B-spline-based generative adversarial network for shape parameterization and a mixture of multilayer perceptron, recurrent neural networks, and experts for surrogate modeling. The framework's optimization results align well with those from direct CFD-based optimizations and evaluations for both subsonic and transonic conditions.

AEROSPACE SCIENCE AND TECHNOLOGY (2021)

Article Computer Science, Interdisciplinary Applications

A conservative multi-fidelity surrogate model-based robust optimization method for simulation-based optimization

Jiexiang Hu et al.

Summary: The paper proposed a new multi-fidelity surrogate model-based robust optimization method to address prediction uncertainty, achieving better optimal design solutions.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2021)

Article Computer Science, Interdisciplinary Applications

Emulator embedded neural networks for multi-fidelity conceptual design exploration of hypersonic vehicles

Atticus Beachy et al.

Summary: By embedding multi-fidelity models in selected neurons of hidden layers, the proposed method accelerates training, reduces reliance on high-fidelity data, decreases data sampling, improves accuracy, lowers overfitting risk, and achieves robust fitting with randomly initialized weights.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2021)

Article Computer Science, Interdisciplinary Applications

A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems

Xuhui Meng et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2020)

Article Computer Science, Interdisciplinary Applications

Toward the robust establishment of variable-fidelity surrogate models for hierarchical stiffened shells by two-step adaptive updating approach

Kuo Tian et al.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2020)

Article Computer Science, Artificial Intelligence

Deep neural network based framework for complex correlations in engineering metrics

Vahid Asghari et al.

ADVANCED ENGINEERING INFORMATICS (2020)

Article Computer Science, Interdisciplinary Applications

Scalable gradient-enhanced artificial neural networks for airfoil shape design in the subsonic and transonic regimes

Mohamed Amine Bouhlel et al.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2020)

Article Computer Science, Interdisciplinary Applications

Data-driven modelling and optimization of stiffeners on undevelopable curved surfaces

Kuo Tian et al.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2020)

Article Engineering, Mechanical

A Kriging Surrogate Model for Computing Gas Mixture Equations of State

Frederick Ouellet et al.

JOURNAL OF FLUIDS ENGINEERING-TRANSACTIONS OF THE ASME (2019)

Article Computer Science, Artificial Intelligence

A comparison of deep networks with ReLU activation function and linear spline-type methods

Konstantin Eckle et al.

NEURAL NETWORKS (2019)

Article Computer Science, Interdisciplinary Applications

An active learning reliability method with multiple kernel functions based on radial basis function

Lingjian Shi et al.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2019)

Article Computer Science, Artificial Intelligence

Multi-objective optimization design of a complex building based on an artificial neural network and performance evaluation of algorithms

Binghui Si et al.

ADVANCED ENGINEERING INFORMATICS (2019)

Article Computer Science, Artificial Intelligence

Surrogate-guided differential evolution algorithm for high dimensional expensive problems

Xiwen Cai et al.

SWARM AND EVOLUTIONARY COMPUTATION (2019)

Article Computer Science, Artificial Intelligence

Gaussian Process Surrogate Models for the CMA Evolution Strategy

Lukas Bajer et al.

EVOLUTIONARY COMPUTATION (2019)

Article Engineering, Multidisciplinary

An active learning method combining Kriging and accelerated chaotic single loop approach (AK-ACSLA) for reliability-based design optimization

Zeng Meng et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2019)

Review Computer Science, Interdisciplinary Applications

A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design

Haitao Liu et al.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2018)

Article Engineering, Industrial

A new adaptive sequential sampling method to construct surrogate models for efficient reliability analysis

Ning-Cong Xiao et al.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2018)

Article Computer Science, Interdisciplinary Applications

A space mapping method based on Gaussian process model for variable fidelity metamodeling

Ping Jiang et al.

SIMULATION MODELLING PRACTICE AND THEORY (2018)

Article Computer Science, Artificial Intelligence

A sequential multi-fidelity metamodeling approach for data regression

Qi Zhou et al.

KNOWLEDGE-BASED SYSTEMS (2017)

Article Engineering, Aerospace

Influences of shield ratio on the infrared signature of serpentine nozzle

Wen Cheng et al.

AEROSPACE SCIENCE AND TECHNOLOGY (2017)

Article Computer Science, Interdisciplinary Applications

Kriging-based optimization design for a new style shell with black box constraints

Huachao Dong et al.

JOURNAL OF ALGORITHMS & COMPUTATIONAL TECHNOLOGY (2017)

Article Computer Science, Artificial Intelligence

A variable fidelity information fusion method based on radial basis function

Qi Zhou et al.

ADVANCED ENGINEERING INFORMATICS (2017)

Article Computer Science, Artificial Intelligence

An active learning metamodeling approach by sequentially exploiting difference information from variable-fidelity models

Qi Zhou et al.

ADVANCED ENGINEERING INFORMATICS (2016)

Article Computer Science, Artificial Intelligence

Transfer learning using computational intelligence: A survey

Jie Lu et al.

KNOWLEDGE-BASED SYSTEMS (2015)

Article Engineering, Mechanical

Trust Region Based Mode Pursuing Sampling Method for Global Optimization of High Dimensional Design Problems

George H. Cheng et al.

JOURNAL OF MECHANICAL DESIGN (2015)

Article Computer Science, Artificial Intelligence

A prior-knowledge input LSSVR metamodeling method with tuning based on cellular particle swarm optimization for engineering design

Jun Zheng et al.

EXPERT SYSTEMS WITH APPLICATIONS (2014)

Article Engineering, Multidisciplinary

RECURSIVE CO-KRIGING MODEL FOR DESIGN OF COMPUTER EXPERIMENTS WITH MULTIPLE LEVELS OF FIDELITY

Loic Le Gratiet et al.

INTERNATIONAL JOURNAL FOR UNCERTAINTY QUANTIFICATION (2014)

Article Engineering, Aerospace

Improving variable-fidelity surrogate modeling via gradient-enhanced kriging and a generalized hybrid bridge function

Zhong-Hua Han et al.

AEROSPACE SCIENCE AND TECHNOLOGY (2013)

Article Computer Science, Interdisciplinary Applications

SO-MI: A surrogate model algorithm for computationally expensive nonlinear mixed-integer black-box global optimization problems

Juliane Muller et al.

COMPUTERS & OPERATIONS RESEARCH (2013)

Article Automation & Control Systems

Active learning for spectroscopic data regression

Fouzi Douak et al.

JOURNAL OF CHEMOMETRICS (2012)

Article Operations Research & Management Science

Expected improvement in efficient global optimization through bootstrapped kriging

Jack P. C. Kleijnen et al.

JOURNAL OF GLOBAL OPTIMIZATION (2012)

Article Computer Science, Artificial Intelligence

A Survey on Transfer Learning

Sinno Jialin Pan et al.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2010)

Article Computer Science, Hardware & Architecture

Accurate and efficient processor performance prediction via regression tree based modeling

Bin Li et al.

JOURNAL OF SYSTEMS ARCHITECTURE (2009)

Article Computer Science, Interdisciplinary Applications

Ensemble of surrogates

Tushar Goel et al.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2007)