428 Hits in 4.8 sec

Data-Driven Evolutionary Algorithm With Perturbation-Based Ensemble Surrogates

Jian-Yu Li, Zhi-Hui Zhan, Hua Wang, Jun Zhang
2020 IEEE Transactions on Cybernetics  
Data-driven evolutionary algorithms (DDEAs) aim to utilize data and surrogates to drive optimization, which is useful and efficient when the objective function of the optimization problem is expensive  ...  For the experimental comparisons, a specific DDEA-PES algorithm is developed as an instance by adopting a genetic algorithm as the optimizer and radial basis function neural networks as the base models  ...  Furthermore, based on a set of surrogates, committee-based active learning for the surrogate-assisted PSO algorithm (CAL-SAPSO) employed the committee-based decision for predictions [39] .  ... 
doi:10.1109/tcyb.2020.3008280 pmid:32776886 fatcat:ntefcneom5bzhoyl37ixewio6m

Comparative Study of Surrogate Modeling Methods for Signal Integrity and Microwave Circuit Applications

Thong Nguyen, Bobi Shi, Hanzhi Ma, Er-Ping Li, Xu Chen, Andreas C. Cangellaris, Jose Schutt-Aine
2021 IEEE Transactions on Components, Packaging, and Manufacturing Technology  
Furthermore, increased complexity in circuitry and integration compounds design iteration and the associated, high-dimensional sensitivity analysis and performance optimization studies.  ...  Three design applications, a high-speed channel, a millimeter-wave filter, and a low-noise amplifier are used to demonstrate the robustness of the proposed Gaussian Process based surrogate models.  ...  In the quest for computationally efficient methods capable of handling the high-dimensional design space of such devices and systems, machine learning (ML) methods are being explored recently for modeling  ... 
doi:10.1109/tcpmt.2021.3098666 fatcat:axvxczao5bdp5locslzl4ectne

Support Vector Regression based Active Subspace (SVR-AS) Modeling of High-Speed Links for Fast and Accurate Sensitivity Analysis

Hanzhi Ma, Er-Ping Li, Andreas C. Cangellaris, Xu Chen
2020 IEEE Access  
A methodology based on the joint usage of support vector regression and active subspace is introduced in this paper for accelerated sensitivity analysis of high-speed links through parameter space dimensionality  ...  The resulting reduced-dimensional model is shown to perform well in sensitivity analysis of high-speed links including IBIS-AMI equalization, and is computationally more efficient than Sobol's method.  ...  However, Sobol's method using Monte Carlo integral is computationally expensive, especially when the dimensionality of the space of input parameters is high.  ... 
doi:10.1109/access.2020.2988088 fatcat:pnip7v3bwjaynkymulmsnb6nte

Accurate Modeling of Frequency Selective Surfaces Using Fully-Connected Regression Model With Automated Architecture Determination and Parameter Selection Based on Bayesian Optimization

Nurullah Calik, Mehmet Ali Belen, Peyman Mahouti, Slawomir Koziel
2021 IEEE Access  
A common practice is experience-driven setup, heavily based on trial and error, which does not guarantee the optimum model determination and may lead to multiple problems such as poor generalization or  ...  Surrogate modeling has become an important tool in the design of high-frequency structures.  ...  ACKNOWLEDGMENT The authors would like to thank Dassault Systemes, France, for making CST Microwave Studio available and Signal Processing for Computational Intelligence Group in Informatics Institute of  ... 
doi:10.1109/access.2021.3063523 fatcat:bolinrbk55dsjmjwau6pwyupgy

Machine Learning in Heterogeneous Porous Materials [article]

Marta D'Elia, Hang Deng, Cedric Fraces, Krishna Garikipati, Lori Graham-Brady, Amanda Howard, George Karniadakis, Vahid Keshavarzzadeh, Robert M. Kirby, Nathan Kutz, Chunhui Li, Xing Liu (+12 others)
2022 arXiv   pre-print
areas of heterogeneous materials, machine learning (ML) and applied mathematics to identify how ML can advance materials research.  ...  The "Workshop on Machine learning in heterogeneous porous materials" brought together international scientific communities of applied mathematics, porous media, and material sciences with experts in the  ...  We would also like to thank the Department of Mechanical Engineering at The University of Utah for assistance with the logistics and behind the scene preparations for the workshop.  ... 
arXiv:2202.04137v1 fatcat:tuhghvcifnebzo2pcwifeek4vu

Surrogate regression modelling for fast seismogram generation and detection of microseismic events in heterogeneous velocity models

Saptarshi Das, Xi Chen, Michael P Hobson, Suhas Phadke, Bertwim van Beest, Jeroen Goudswaard, Detlef Hohl
2018 Geophysical Journal International  
Surrogate meta-models or proxy methods were traditionally developed for various optimization problems e.g. constrained single or multi-objective optimization problems, missing data problems etc.  ...  Such a statistical learning or approximation of physics in the form of PDE solver's outputs has been widely used in various surrogate meta-model assisted optimization methods before e.g. in (Forrester  ...  Acknowledgement This work has been supported by the Shell Projects and Technology. The Wilkes high performance GPU computing service at the University of Cambridge has been used in this work.  ... 
doi:10.1093/gji/ggy283 fatcat:shjrbiinrjdqtkkohdyddr54a4

Integrating machine learning and multiscale modeling—perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences

Mark Alber, Adrian Buganza Tepole, William R. Cannon, Suvranu De, Salvador Dura-Bernal, Krishna Garikipati, George Karniadakis, William W. Lytton, Paris Perdikaris, Linda Petzold, Ellen Kuhl
2019 npj Digital Medicine  
However, machine learning alone ignores the fundamental laws of physics and can result in ill-posed problems or non-physical solutions.  ...  , and inform decision making for the benefit of human health.  ...  , low fidelity and sparse, expensive, high fidelity data from experiments and simulations to create efficient and robust surrogate models.  ... 
doi:10.1038/s41746-019-0193-y pmid:31799423 pmcid:PMC6877584 fatcat:uhgdhq7rjffqnboydb3e6d2tuu

The Future of Sensitivity Analysis: An Essential Discipline for Systems Modeling and Policy Support

Saman Razavi, Anthony Jakeman, Andrea Saltelli, Clémentine Prieur, Bertrand Iooss, Emanuele Borgonovo, Elmar Plischke, Samuele Lo Piano, Takuya Iwanaga, William Becker, Stefano Tarantola, Joseph H.A. Guillaume (+14 others)
2020 Environmental Modelling & Software  
applications to solve real-world problems.  ...  computational burden of SA, (4) progressing SA in the context of machine learning, (5) clarifying the relationship and role of SA to uncertainty quantification, and (6) evolving the use of SA in support  ...  We are thankful to the sponsors of this event, including the French research association on stochastic methods for the analysis of numerical codes (MASCOT-NUM), Open Evidence Research at Universitat Oberta  ... 
doi:10.1016/j.envsoft.2020.104954 fatcat:7uwrzzb4cjc4lh4h4uqqznsfi4

A Multi-Facet Survey on Memetic Computation

Xianshun Chen, Yew-Soon Ong, Meng-Hiot Lim, Kay Chen Tan
2011 IEEE Transactions on Evolutionary Computation  
computation, multiagent system, multiobjective memetic algorithms, surrogate-assisted memetic algorithms.  ...  Memetic computation is a paradigm that uses the notion of meme(s) as units of information encoded in computational representations for the purpose of problem-solving.  ...  computationally expensive problems.  ... 
doi:10.1109/tevc.2011.2132725 fatcat:dy4vpyft6rhdxb4dkt7iyaqvna

Safe Model-based Off-policy Reinforcement Learning for Eco-Driving in Connected and Automated Hybrid Electric Vehicles [article]

Zhaoxuan Zhu, Nicola Pivaro, Shobhit Gupta, Abhishek Gupta, Marcello Canova
2022 arXiv   pre-print
The eco-driving problem seeks to design optimal speed and power usage profiles based upon look-ahead information from connectivity and advanced mapping features.  ...  While the previous studies synthesize simulators and model-free DRL to reduce online computation, this work proposes a Safe Off-policy Model-Based Reinforcement Learning algorithm for the eco-driving problem  ...  While the trajectory optimization increases the sample efficiency and helps exploration [18] , solving the trajectory optimization problem makes each data sample more computationally expensive.  ... 
arXiv:2105.11640v2 fatcat:6nj7e5vej5b3rkfp4snjk7nt4e

Machine Learning in Nano-Scale Biomedical Engineering [article]

Alexandros-Apostolos A. Boulogeorgos, Stylianos E. Trevlakis, Sotiris A. Tegos, Vasilis K. Papanikolaou, George K. Karagiannidis
2020 arXiv   pre-print
Machine learning (ML) empowers biomedical systems with the capability to optimize their performance through modeling of the available data extremely well, without using strong assumptions about the modeled  ...  For each of the presented methodologies, special emphasis is given to its principles, applications, and limitations.  ...  Similarly to regression, SVMs are efficient methods for problems with high-dimensional spaces.  ... 
arXiv:2008.02195v2 fatcat:5i445iipdnag3pqukyeq2ceopy

Machine Learning in Chemical Product Engineering: The State of the Art and a Guide for Newcomers

Cindy Trinh, Dimitrios Meimaroglou, Sandrine Hoppe
2021 Processes  
In recent years, artificial intelligence (AI) and machine learning (ML) methods have gained increasing attention due to their performance in tackling particularly complex problems in various areas, such  ...  This review is further completed by general guidelines for the selection of an appropriate ML technique given the characteristics of each problem and by a critical discussion of several key issues associated  ...  Method Advantages Limitations Physical-based • Accurate and generalizable • Deep understanding of chemistry • Not adapted for high-throughput reaction prediction tasks and large systems (computationally  ... 
doi:10.3390/pr9081456 fatcat:tb2fredhnjghhm7lsj5qb2bipq

Taming an autonomous surface vehicle for path following and collision avoidance using deep reinforcement learning [article]

Eivind Meyer, Haakon Robinson, Adil Rasheed, Omer San
2019 arXiv   pre-print
In this article, we explore the feasibility of applying proximal policy optimization, a state-of-the-art deep reinforcement learning algorithm for continuous control tasks, on the dual-objective problem  ...  based on the OpenAI gym python toolkit.  ...  Acknowledgment The authors acknowledge the financial support from the Norwegian Research Council and the industrial partners DNV GL, Kongsberg and Maritime Robotics of the Autosit project.  ... 
arXiv:1912.08578v1 fatcat:zjnce37f3fau3jmkkspaipcyh4

Review of Physics-based and Data-driven Multiscale Simulation Methods for Computational Fluid Dynamics and Nuclear Thermal Hydraulics [article]

Arsen S. Iskhakov, Nam T. Dinh
2021 arXiv   pre-print
Possible applications of the data-driven (machine learning and statistics-based) methods for enabling and / or improving multiscale modeling (bridging the scaling gaps) are reviewed.  ...  expensive.  ...  ML can also assist in learning non-linear and / or stochastic PDEs (e.g., (Raissi & Karniadakis, 2018) ).  ... 
arXiv:2102.01159v1 fatcat:foydy65exnbonjeswdxittimie

DCBT-Net: Training Deep Convolutional Neural Networks with Extremely Noisy Labels

Bekhzod Olimov, Jeonghong Kim, Anand Paul
2020 IEEE Access  
ACKNOWLEDGMENT This research was conducted using the assistance of the "HPC Support" Project, supported by the 'Ministry of Science and ICT' and NIPA.  ...  and computationally expensive.  ...  Since this dataset comprises of significantly large number of images for training, the experiments on the dataset were time-consuming and computationally expensive.  ... 
doi:10.1109/access.2020.3041873 fatcat:gxnadiyc7vadvnpmknqya2yfri
« Previous Showing results 1 — 15 out of 428 results