Multi-objective optimization of a stacked neural network using an evolutionary hyper-heuristic

Furtuna, Renata; Curteanu, Silvia; Leon, Florin

doi:10.1016/j.asoc.2011.09.001

Cited by 46 publications

(21 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NSGAII learned to choose from a set of rules representing a constructive heuristic for 2D irregular stock cutting. In (Furtuna et al, 2012) a multi-objective hyper-heuristic for the design and optimization of a stacked neural network is proposed. The proposed approach is based on NSGAII combined with a local search algorithm (Quasi-Newton algorithm).…”

Section: Introductionmentioning

confidence: 99%

A multi-objective hyper-heuristic based on choice function

Maashi

Özcan

Kendall

2014

Expert Systems with Applications

View full text Add to dashboard Cite

Hyper-heuristics are emerging methodologies that perform a search over the space of heuristics in an attempt to solve difficult computational optimization problems. We present a learning selection choice function based hyper-heuristic to solve multi-objective optimization problems. This high level approach controls and combines the strengths of three well-known multi-objective evolutionary algorithms (i.e. NSGAII, SPEA2 and MOGA), utilizing them as the low level heuristics. The performance of the proposed learning hyper-heuristic is investigated on the Walking Fish Group test suite which is a common benchmark for multi-objective optimization. Additionally, the proposed hyper-heuristic is applied to the vehicle crashworthiness design problem as a real-world multi-objective problem. The experimental results demonstrate the effectiveness of the hyper-heuristic approach when compared to the performance of each low level heuristic run on its own, as well as being compared to other approaches including an adaptive multi-method search, namely AMALGAM.

show abstract

Section: Introductionmentioning

confidence: 99%

A multi-objective hyper-heuristic based on choice function

Maashi

Özcan

Kendall

2014

Expert Systems with Applications

View full text Add to dashboard Cite

show abstract

“…Hierarchical genetic algorithms, which used parametric and control genes to construct the chromosome, were applied for neuroevolution by Elhachmi and Guennoun [126]. On the side of ANN training procedures the focus is in recent years on novel combinations of GA with gradient-based or local optimization methods, which were used to address the problem of stock market time-series prediction [120] and optimize multi-objective processes in material synthesis [128].…”

Section: Hybrid Ann+easmentioning

confidence: 99%

Adaptation and Hybridization in Nature-Inspired Algorithms

Fister

Strnad

Yang³

2015

Adaptation, Learning, and Optimization

View full text Add to dashboard Cite

Abstract. The aim of this chapter is to familiarize readers with the basics of adaptation and hybridization in nature-inspired algorithms as necessary for understanding the main contents of this book. Adaptation is a metaphor for flexible autonomous systems that respond to external changing factors (mostly environmental) by adapting their wellestablished behavior. Adaptation emerges in practically all areas of human activities as well. Such adaptation mechanisms can be used as a general problem-solving approach, though it may suffer from a lack of problem-specific knowledge. To solve specific problems with additional improvements of possible performance, hybridization can be used in order to incorporate a problem-specific knowledge from a problem domain. In order to discuss relevant issues as general as possible, the classification of problems is identified at first. Additionally, we focus on the biological foundations of adaptation that constitute the basis for the formulation of nature-inspired algorithms. This book highlights three types of inspirations from nature: the human brain, Darwinian natural selection, and the behavior of social living insects (e.g., ants, bees, etc.) and animals (e.g., swarm of birds, shoals of fish, etc.), which influence the development of artificial neural networks. evolutionary algorithms, and swarm intelligence, respectively. The mentioned algorithms that can be placed under the umbrella of computational intelligence are described from the viewpoint of adaptation and hybridization so as to show that these mechanisms are simple to develop and yet very efficient. Finally, a brief review of recent developed applications is presented.

show abstract

“…For MWO, value of the shape parameter of the levy flight of 1.8, short range reference coefficient of 0.11, long range reference coefficient of 0.75, moving coefficient of 0.63, 1.26 and 1.05, walking scale of 0.1, Levy distribution of 1.5, and scale factor of space of 0.5 are selected [63]. For lower-level training of the soft TDNN sensor, sigmoid transformation function, fully complex architecture, Levenberg-Marquardt training method, epochs number of 100, learning rate of 0.1, and the range of weights and thresholds of [-1,1] are selected [69]. It should be also mentioned that all of the hyper-level BIC architecture evolving algorithms continue the optimization for 1000 function evaluations.…”

Section: Parameter Selection and Performance Metrics Verificationmentioning

confidence: 99%

Biologically inspired time-delay soft sensors for online monitoring of automotive coldstart operations: a comparative analysis

Azad

Mozaffari

2015

Meccanica

View full text Add to dashboard Cite

In this paper, an attempt is made to develop a bio-inspired soft sensor as a real-time identifier for online estimations of the engine-out hydrocarbon emission and exhaust gas temperature over the coldstart operation of an automotive engine. To date, a number of studies have been done to design soft models as offline identifiers for capturing the knowledge of a set of collected coldstart databases. The results indicated that bio-inspired computation can be of great use for the identification of coldstart phenomenon. It also can be a good alternative to classic approaches. However, bio-inspired soft sensors have not been used as online estimators for monitoring the behavior of coldstart phenomenon. This is while it is desirable to have a flexible, inexpensive online identifier which can be employed to detect the realtime behavior of the engine over coldstart periods.Here, an exhaustive numerical study is conducted to demonstrate the potentials of bio-inspired approaches for online measuring of engine-out hydrocarbon emission and exhaust gas temperature during the coldstart operation. To do so, a set of well-known bioinspired optimizers are combined with a time-delayed artificial neural network to design an evolvable soft sensor. The results clearly demonstrate the high potential of bio-inspired computation for handling the difficulties associated with the real-time identification of engine-out hydrocarbon emission and exhaust gas temperature.

show abstract

Multi-objective optimization of a stacked neural network using an evolutionary hyper-heuristic

Cited by 46 publications

References 26 publications

A multi-objective hyper-heuristic based on choice function

A multi-objective hyper-heuristic based on choice function

Adaptation and Hybridization in Nature-Inspired Algorithms

Biologically inspired time-delay soft sensors for online monitoring of automotive coldstart operations: a comparative analysis

Contact Info

Product

Resources

About