Experience-based Computation: Learning to Optimise

Informations projet

ECOLE

N° de convention de subvention: 766186

DOI

10.3030/766186

Projet clôturé

Date de signature de la CE 10 Août 2017

Date de début 1 Avril 2018

Date de fin 31 Mars 2022

Financé au titre de

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Coût total

€ 2 060 348,04

Contribution de l’UE

€ 2 060 348,04

Coordonné par

THE UNIVERSITY OF BIRMINGHAM
United Kingdom

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Demonstrators, pilots, prototypes (2)

Summer schools & Career events

D6.1, D6.2, D6.3, D6.4 D6.5 Project website, Dissemination & outreach plan, Summer schools and Career events

Dissemination and outreach plan

D6.1, D6.2, D6.3, D6.4 D6.5 Project website, Dissemination & outreach plan, Summer schools and Career events

Documents, reports (17)

Engineering data and descriptors

D1.1, D1.2, D1.3 software tools, benchmark test suite of automotive multi-disciplinary/domain real-world scenario

Experience-based high dimensional & big data assisted optimisation

D21 D21 D23 D24 software tools for high dimensional big data assisted optimisation Constrained Multicriteria optimisation Robustness Uncertainty Modelling in optimisation

Multi-domain optimisation software package based on online experience exploitation in dynamic environments

D11 D12 D13 software tools benchmark test suite of automotive multidisciplinarydomain realworld scenario

Constrained & Multicriteria optimisation

D21 D21 D23 D24 software tools for high dimensional big data assisted optimisation Constrained Multicriteria optimisation Robustness Uncertainty Modelling in optimisation

Robustness & Uncertainty Modelling in experience-based optimisation

D21 D21 D23 D24 software tools for high dimensional big data assisted optimisation Constrained Multicriteria optimisation Robustness Uncertainty Modelling in optimisation

Research and personal skill report, Final

D41 D42 D43 ECOLE training programme Research and personal skill reports Career development reports

Deep structured learning and model space learning for engineering and ICT data

D31 D32 D33 D34 D35 software tools on semisupervised learning for class imbalance problems model space learning and representation learning text mining with deep probabilistic models learning for proactive dynamic and robust optimisation with online feature selection

Dissemination and outreach final report

D61 D62 D63 D64 D65 Project website Dissemination outreach plan Summer schools and Career events

Dissemination and outreach mid-term plan & report

D6.1, D6.2, D6.3, D6.4 D6.5 Project website, Dissemination & outreach plan, Summer schools and Career events

Text mining models for product feature optimisation

D31 D32 D33 D34 D35 software tools on semisupervised learning for class imbalance problems model space learning and representation learning text mining with deep probabilistic models learning for proactive dynamic and robust optimisation with online feature selection

Multi-criteria optimisation software environment based on learning for adaptive feature selection and constraint prediction

D11 D12 D13 software tools benchmark test suite of automotive multidisciplinarydomain realworld scenario

Semi-supervised learning for class imbalance problems

D31 D32 D33 D34 D35 software tools on semisupervised learning for class imbalance problems model space learning and representation learning text mining with deep probabilistic models learning for proactive dynamic and robust optimisation with online feature selection

Integrated software environment (‘Self-Tuning optimisation) and manual

D21 D21 D23 D24 software tools for high dimensional big data assisted optimisation Constrained Multicriteria optimisation Robustness Uncertainty Modelling in optimisation

Integrated software environment and manual

D31 D32 D33 D34 D35 software tools on semisupervised learning for class imbalance problems model space learning and representation learning text mining with deep probabilistic models learning for proactive dynamic and robust optimisation with online feature selection

Research and personal skill report, mid-term

D4.1, D4.2, D4.3 ECOLE training programme, Research and personal skill reports, Career development reports

Online learning for proactive dynamic and robust optimization

D31 D32 D33 D34 D35 software tools on semisupervised learning for class imbalance problems model space learning and representation learning text mining with deep probabilistic models learning for proactive dynamic and robust optimisation with online feature selection

Career development report, yearly

Career development report yearly

Other (2)

Awarding of doctoral degrees
ECOLE-training programme

D4.1, D4.2, D4.3 ECOLE training programme, Research and personal skill reports, Career development reports

Publications

Conference proceedings (30)

A new acquisition function for robust Bayesian optimization of unconstrained problems

Auteurs: Sibghat Ullah, Hao Wang, Stefan Menzel, Bernhard Sendhoff, Thomas Bäck
Publié dans: 2021 Genetic and Evolutionary Computation Conference, GECCO 2021, 2021, ISBN 9781450383516
Éditeur: Association for Computing Machinery
DOI: 10.1145/3449726.3463206

Point2FFD: Learning Shape Representations of Simulation-Ready 3D Models for Engineering Design Optimization

Auteurs: Thiago Rios, Bas Van Stein, Thomas Bäck, Bernhard Sendhoff, Stefan Menzel
Publié dans: 2021 International Conference on 3D Vision (3DV), 2021, ISBN 978-1-6654-2689-3
Éditeur: IEEE
DOI: 10.1109/3dv53792.2021.00110

Exploring Clinical Time Series Forecasting with Meta-Features in Variational Recurrent Models

Auteurs: Sibghat Ullah, Zhao Xu, Hao Wang, Stefan Menzel, Bernhard Sendhoff, Thomas Back
Publié dans: 2020 International Joint Conference on Neural Networks (IJCNN), 2020, Page(s) 1-9, ISBN 978-1-7281-6926-2
Éditeur: IEEE
DOI: 10.1109/ijcnn48605.2020.9207254

Artificial Neural Networks as Feature Extractors in Continuous Evolutionary Optimization

Auteurs: Stephen Friess,  Peter Tiňo; Zhao Xu; Stefan Menzel; Bernhard Sendhoff; Xin Yao
Publié dans: 2021 International Joint Conference on Neural Networks (IJCNN), 2021, ISBN 978-1-6654-4597-9
Éditeur: IEEE
DOI: 10.1109/ijcnn52387.2021.9533915

Exploring Dimensionality Reduction Techniques for Efficient Surrogate-Assisted Optimization

Auteurs: Sibghat Ullah, Duc Anh Nguyen, Hao Wang, Stefan Menzel, Bernhard Sendhoff and Thomas Bäck
Publié dans: IEEE Symposium Series on Computational Intelligence (SSCI), 2020
Éditeur: IEEE

Quantifying The Generative Capabilities Of Variational Autoencoders For 3D Car Point Clouds

Auteurs: Sneha Saha, Stefan Menzel, Leandro Minku, Xin Yao, Bernhard Sendhoff and Patricia Wollstadt
Publié dans: 2020 IEEE Symposium Series on Computational Intelligence (SSCI), 2020
Éditeur: IEEE

Exploiting Generative Models for Performance Predictions of 3D Car Designs

Auteurs: Sneha Saha, Thiago Rios, Leandro Minku, Bas Vas Stein, Patricia Wollstadt, Xin Yao, Thomas Back, Bernhard Sendhoff, Stefan Menzel
Publié dans: 2021 IEEE Symposium Series on Computational Intelligence (SSCI), 2022, ISBN 978-1-7281-9049-5
Éditeur: IEEE
DOI: 10.1109/ssci50451.2021.9660034

Spatio-Temporal Activity Recognition for Evolutionary Search Behavior Prediction

Auteurs: S. Friess, P. Tiňo, S. Menzel, Z. Xu, B. Sendhoff and X. Yao
Publié dans: 2022 International Joint Conference on Neural Networks (IJCNN), 2022
Éditeur: IEEE

Exploiting 3D Variational Autoencoders For Interactive Vehicle Design

Auteurs: S. Saha, L. L. Minku, X. Yao, B. Sendhoff , and S. Menzel
Publié dans: 2022 International Design Conference, 2022
Éditeur: Cambridge Press
DOI: 10.1017/pds.2022.177

Feature Visualization for 3D Point Cloud Autoencoders

Auteurs: Thiago Rios, Bas van Stein, Stefan Menzel, Thomas Back, Bernhard Sendhoff, Patricia Wollstadt
Publié dans: 2020 International Joint Conference on Neural Networks (IJCNN), 2020, Page(s) 1-9, ISBN 978-1-7281-6926-2
Éditeur: IEEE
DOI: 10.1109/ijcnn48605.2020.9207326

Learning Sparsity of Representations with Discrete Latent Variables

Auteurs: Zhao Xu,  Daniel Onoro Rubio, Giuseppe Serra, Mathias Niepert
Publié dans: 2021 International Joint Conference on Neural Networks (IJCNN), 2021, ISBN 978-1-6654-4597-9
Éditeur: IEEE
DOI: 10.1109/ijcnn52387.2021.9533762

Interpreting Node Embedding with Text-labeled Graphs

Auteurs: Giuseppe Serra, Zhao Xu, Mathias Niepert, Carolin Lawrence, Peter Tiňo, Xin Yao
Publié dans: 2021 International Joint Conference on Neural Networks (IJCNN), 2021, ISBN 978-1-6654-4597-9
Éditeur: IEEE
DOI: 10.1109/ijcnn52387.2021.9533692

Learning Time-Series Data of Industrial Design Optimization using Recurrent Neural Networks

Auteurs: Sneha Saha, Thiago Rios, Stefan Menzel, Bernhard Sendhoff, Thomas Back, Xin Yao, Zhao Xu, Patricia Wollstadt
Publié dans: 2019 International Conference on Data Mining Workshops (ICDMW), 2019, Page(s) 785-792, ISBN 978-1-7281-4896-0
Éditeur: IEEE
DOI: 10.1109/icdmw.2019.00116

Optimal Evolutionary Optimization Hyper-parameters to Mimic Human User Behavior

Auteurs: Sneha Saha, Thiago Rios, Leandro L. Minku, Xin Yao, Zhao Xu, Bernhard Sendhoff, Stefan Menzel
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 858-866, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9002958

Learning Transferable Variation Operators in a Continuous Genetic Algorithm

Auteurs: Stephen Friess, Peter Tino, Stefan Menzel, Bernhard Sendhoff, Xin Yao
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 2027-2033, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9002976

When and How to Transfer Knowledge in Dynamic Multi-objective Optimization

Auteurs: Gan Ruan, Leandro L. Minku, Stefan Menzel, Bernhard Sendhoff, Xin Yao
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 2034-2041, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9002815

On the Efficiency of a Point Cloud Autoencoder as a Geometric Representation for Shape Optimization

Auteurs: Thiago Rios, Bernhard Sendhoff, Stefan Menzel, Thomas Back, Bas van Stein
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 791-798, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9003161

Hyperparameter Optimisation for Improving Classification under Class Imbalance

Auteurs: Jiawen Kong, Wojtek Kowalczyk, Duc Anh Nguyen, Thomas Back, Stefan Menzel
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 3072-3078, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9002679

Scalability of Learning Tasks on 3D CAE Models Using Point Cloud Autoencoders

Auteurs: Thiago Rios, Patricia Wollstadt, Bas van Stein, Thomas Back, Zhao Xu, Bernhard Sendhoff, Stefan Menzel
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 1367-1374, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9002982

An Empirical Comparison of Meta-Modeling Techniques for Robust Design Optimization

Auteurs: Sibghat Ullah, Hao Wang, Stefan Menzel, Bernhard Sendhoff, Thomas Back
Publié dans: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019, Page(s) 819-828, ISBN 978-1-7281-2485-8
Éditeur: IEEE
DOI: 10.1109/ssci44817.2019.9002805

Predicting CMA-ES Operators as Inductive Biases for Shape Optimization Problems

Auteurs: Stephen Friess, Peter Tiňo, Stefan Menzel, Bernhard Sendhoff, Xin Yao
Publié dans: 2021 IEEE Symposium Series on Computational Intelligence (SSCI), 2021, ISBN 978-1-7281-9049-5
Éditeur: IEEE
DOI: 10.1109/ssci50451.2021.9660001

Exploiting Linear Interpolation of Variational Autoencoders for Satisfying Preferences in Evolutionary Design Optimization

Auteurs: Sneha Saha; Leandro L. Minku; Xin Yao; Bernhard Senhoff; Stefan Menzel
Publié dans: 2021 IEEE Congress on Evolutionary Computation (CEC), 2021, ISBN 978-1-7281-8394-7
Éditeur: IEEE
DOI: 10.1109/cec45853.2021.9504772

Representing Experience in Continuous Evolutionary optimisation through Problem-tailored Search Operators

Auteurs: Stephen Friess, Peter Tino, Stefan Menzel, Bernhard Sendhoff, Xin Yao
Publié dans: 2020 IEEE Congress on Evolutionary Computation (CEC), 2020, Page(s) 1-7, ISBN 978-1-7281-6929-3
Éditeur: IEEE
DOI: 10.1109/cec48606.2020.9185687

Efficient AutoML via Combinational Sampling

Auteurs: Duc Anh Nguyen, Anna V. Kononova, Stefan Menzel, Bernhard Sendhoff, Thomas Back
Publié dans: 2021 IEEE Symposium Series on Computational Intelligence (SSCI), 2021, ISBN 978-1-7281-9049-5
Éditeur: IEEE
DOI: 10.1109/ssci50451.2021.9660073

Split-AE: An Autoencoder-based Disentanglement Framework for 3D Shape-to-shape Feature Transfer

Auteurs: Sneha Saha, Leandro Minku, Xin Yao, Bernard Sendhoff, Stefan Menzel
Publié dans: 2022 International Joint Conference on Neural Networks (IJCNN), 2022
Éditeur: IEEE

Back to Meshes: Optimal Simulation-ready mesh prototypes for autoencoder based 3D car point clouds

Auteurs: Thiago Rios, Jiawen Kong, Bas van Stein, Thomas Bäck, Patricia Wollstadt, Bernhard Sendhoff and Stefan Menzel
Publié dans: 2020 IEEE Symposium Series on Computational Intelligence SSCI, 2020
Éditeur: IEEE

Improved Sample Type Identification for Multi-Class Imbalanced Classification with Real-World Applications

Auteurs: J. Kong, W. Kowalczyk, K. Jonker, S. Menzel and T. Bäck
Publié dans: the 18th Int. Conference on Data Science (ICDATA’22), 2022
Éditeur: Springer

A Systematic Approach to Analyze the Computational Cost of Robustness in Model-Assisted Robust Optimization

Auteurs: S. Ullah, H. Wang, S. Menzel, B. Sendhoff and T. Bäck
Publié dans: 2022 International Conference on Parallel Problem Solving from Nature, 2022
Éditeur: Springer

Improved Automated CASH Optimization with Tree Parzen Estimators for Class Imbalance Problems

Auteurs: Duc Anh Nguyen, Jiawen Kong, Hao Wang, Stefan Menzel, Bernhard Sendhoff, Anna V. Kononova, Thomas Bäck
Publié dans: 2021 IEEE 8th International Conference on Data Science and Advanced Analytics (DSAA), 2021, ISBN 978-1-6654-2100-3
Éditeur: IEEE
DOI: 10.1109/dsaa53316.2021.9564147

Computational Study on Effectiveness of Knowledge Transfer in Dynamic Multi-objective Optimization

Auteurs: Gan Ruan, Leandro L. Minku, Stefan Menzel, Bernhard Sendhoff, Xin Yao
Publié dans: 2020 IEEE Congress on Evolutionary Computation (CEC), 2020, Page(s) 1-8, ISBN 978-1-7281-6929-3
Éditeur: IEEE
DOI: 10.1109/cec48606.2020.9185907

Book chapters (3)

Improving Sampling in Evolution Strategies Through Mixture-Based Distributions Built from Past Problem Instances

Auteurs: Stephen Friess, Peter Tiňo, Stefan Menzel, Bernhard Sendhoff, Xin Yao
Publié dans: Parallel Problem Solving from Nature – PPSN XVI - 16th International Conference, PPSN 2020, Leiden, The Netherlands, September 5-9, 2020, Proceedings, Part I, Numéro 12269, 2020, Page(s) 583-596, ISBN 978-3-030-58111-4
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-58112-1_40

Improving Imbalanced Classification by Anomaly Detection

Auteurs: Jiawen Kong, Wojtek Kowalczyk, Stefan Menzel, Thomas Bäck
Publié dans: Parallel Problem Solving from Nature – PPSN XVI - 16th International Conference, PPSN 2020, Leiden, The Netherlands, September 5-9, 2020, Proceedings, Part I, Numéro 12269, 2020, Page(s) 512-523, ISBN 978-3-030-58111-4
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-58112-1_35

On the Performance of Oversampling Techniques for Class Imbalance Problems

Auteurs: Jiawen Kong, Thiago Rios, Wojtek Kowalczyk, Stefan Menzel, Thomas Bäck
Publié dans: Advances in Knowledge Discovery and Data Mining - 24th Pacific-Asia Conference, PAKDD 2020, Singapore, May 11–14, 2020, Proceedings, Part II, Numéro 12085, 2020, Page(s) 84-96, ISBN 978-3-030-47435-5
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-47436-2_7

Other (1)

Exploiting Local Geometric Features in Vehicle Design Optimization with 3D Point Cloud Autoencoders

Auteurs: Thiago Rios, Bas van Stein, Patricia Wollstadt, Thomas Bäck, Bernhard Sendhoff, Stefan Menzel
Publié dans: 2021 IEEE Congress on Evolutionary Computation (CEC), 2021
Éditeur: IEEE
DOI: 10.1109/cec45853.2021.9504746

Peer reviewed articles (2)

Multi-Task Shape Optimization Using a 3D Point Cloud Autoencoder as Unified Representation

Auteurs: Thiago Rios, Bas van Stein, Thomas Bäck, Bernhard Sendhoff, Stefan Menzel
Publié dans: IEEE Transactions on Evolutionary Computation , 2021, ISSN 1089-778X
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tevc.2021.3086308

An Efficient Contesting Procedure for AutoML Optimization

Auteurs: D.A. Nguyen, A.V. Kononova, S. Menzel, B. Sendhoff and T. Bäck
Publié dans: IEEE Access, 2022, ISSN 2169-3536
Éditeur: Institute of Electrical and Electronics Engineers Inc.
DOI: 10.1109/access.2022.3192036