Journal Article10.1016/J.SWEVO.2020.100800
Offline data-driven evolutionary optimization based on tri-training
52
TL;DR: In the proposed algorithm, a tri-training algorithm selects candidate solutions with high-confidence fitness prediction to enrich the training data for surrogate models, and is competitive on the problems of up to 500 decision variables.
read more
Abstract: In offline data-driven evolutionary optimization, no real fitness evaluations is allowed during the optimization, making it extremely challenging to build high-quality surrogates on limited amount of data. This is especially true for large-scale optimization problems where typically a large amount of data is needed for constructing reliable surrogate models. To overcome the data deficiency, semi-supervised learning is introduced to the offline data-driven evolutionary optimization process, where tri-training, a co-training variant, is used to update surrogate models. In the proposed algorithm, a tri-training algorithm selects candidate solutions with high-confidence fitness prediction to enrich the training data for surrogate models. The results on benchmark problems show that the proposed algorithm, compared with three most recent offline data-driven optimization algorithms, is competitive on the problems of up to 500 decision variables.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
A Surrogate-Assisted Evolutionary Feature Selection Algorithm With Parallel Random Grouping for High-Dimensional Classification
TL;DR: Wang et al. as mentioned in this paper proposed a surrogate-assisted evolutionary algorithm (SAEA) with parallel random grouping for feature selection problems, in which three main components consist: a constraint-based sampling strategy, which considers the influence of the constraint boundary and the number of selected features.
55
A Surrogate-Assisted Evolutionary Feature Selection Algorithm With Parallel Random Grouping for High-Dimensional Classification
TL;DR: Experimental results show that the proposed SAEA with parallel random grouping for expensive FS problems generally outperforms traditional, ensemble, and evolutionary FS methods on 14 datasets with up to 10 000 features, especially when the required number of real fitness evaluations is limited.
35
A federated data-driven evolutionary algorithm for expensive multi-/many-objective optimization
TL;DR: In this paper, a federated data-driven evolutionary multi-/many-objective optimization algorithm is proposed to leverage federated learning for surrogate construction so that multiple clients collaboratively train a radial-basis function-network as the global surrogate.
Comprehensive review of modeling, structure, and integration techniques of smart buildings in the cyber-physical-social system
Kai Gong,Jianlin Yang,Xu Wang,Chuanwen Jiang,Zhan Xiong,Ming Zhang,Mingxing Guo,Ran Lv,Su-Wen Wang,Shenxi Zhang +9 more
TL;DR: A brief comprehensive survey of the modeling, controlling, and operation of smart buildings is provided, and a novel CPSS-based smart building operation structure is proposed, and the integration techniques for the group ofsmart buildings are discussed.
17
Surrogate-Assisted Environmental Selection for Fast Hypervolume-based Many-Objective Optimization
TL;DR: In this article , the authors proposed an efficient surrogate-assisted greedy inclusion algorithm to deal with computationally expensive hypervolume-based environmental selection (HVES), which aims at selecting a subpopulation with the maximal hypervolume from the current population, plays a crucial role in guiding evolution.
14
References
Efficient Global Optimization of Expensive Black-Box Functions
TL;DR: This paper introduces the reader to a response surface methodology that is especially good at modeling the nonlinear, multimodal functions that often occur in engineering and shows how these approximating functions can be used to construct an efficient global optimization algorithm with a credible stopping rule.
Combining labeled and unlabeled data with co-training
Avrim Blum,Tom M. Mitchell +1 more
- 24 Jul 1998
TL;DR: A PAC-style analysis is provided for a problem setting motivated by the task of learning to classify web pages, in which the description of each example can be partitioned into two distinct views, to allow inexpensive unlabeled data to augment, a much smaller set of labeled examples.
6.4K
Semi-Supervised Learning Literature Survey
Xiaojin Zhu
- 01 Jan 2005
TL;DR: This chapter provides background information on agile principles and an overview of three agile methodologies and an underlying assumption in plan-driven processes is that the requirements are relatively stable.
A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms
TL;DR: The basics are discussed and a survey of a complete set of nonparametric procedures developed to perform both pairwise and multiple comparisons, for multi-problem analysis are given.
4.9K
Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real-Parameter Optimization
Ponnuthurai Nagaratnam Suganthan,Nikolaus Hansen,Jing Liang,Kalyanmoy Deb,Y. P. Chen,Anne Auger,Santosh Tiwari +6 more
- 01 Jan 2005
TL;DR: This special session is devoted to the approaches, algorithms and techniques for solving real parameter single objective optimization without making use of the exact equations of the test functions.