A new efficient training strategy for deep neural networks by hybridization of artificial bee colony and limited–memory BFGS optimization algorithms

Badem, Hasan; Baştürk, Alper; Çalışkan, Abdullah; Yüksel, Mehmet Emin

doi:10.1016/j.neucom.2017.05.061

Cited by 102 publications

(46 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the food source becomes valueless for exploiting no longer, the bees which previously worked at these sources become the scout bees. The scout bees search for the new food sources as randomly Karaboga and Akay (2009), Karaboga and Basturk (2007), Badem et al(2018) Badem et al (2017), Karaboga and Aslan (2016), Karaboga and Aslan (2018), Karaboga and Basturk (2008). The main motivations of the Artificial Bee Colony (ABC) optimization algorithm are the mentioned clever foraging behavior and communication mechanism between honey bees.…”

Section: The Abc Algorithmmentioning

confidence: 99%

“…The amount of the nectar of the food source is also represented to the solution of the fitness value. The employed, scoot and onlooker bees cooperate to optimize the food sources by the iterative manner in the ABC algorithm Karaboga and Akay (2009), Karaboga and Basturk (2007), Badem et al (2017), Karaboga and Aslan (2016), Karaboga and Aslan (2018), Badem et al (2018), Karaboga and Basturk (2008). The fundamental steps of ABC algorithm are presented in Fig.…”

Section: The Abc Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Optimization of the Pool Boiling Heat Transfer in the Region of the Isolated Bubbles using the ABC Algorithm

Alic¹,

Çermik²,

Tokgöz³

et al. 2019

JAFM

View full text Add to dashboard Cite

The region of the isolated bubble regime, in which the bubbling starts, is a very significant process for boiling heat transfer. In this study, Artificial Bee Colony algorithm (ABC), which is mainly based on the searching optimum foods sources for the bees, has been used for the optimization of the pool boiling heat transfer calculation. The ABC algorithm is very handy for numerical analysis. The ABC algorithm has been compared with a genetic algorithm as well as other well-known correlation models for pool boiling heat transfer calculation. The ABC algorithm is found to be useful for any boundary conditions. The boundary conditions have been changed in order to improve the results. Results show that the ABC algorithm works faster than the genetic algorithm for the given problem. The ABC algorithm also predicts less average absolute error when compared with other well-known correlations as well as the optimization using the genetic algorithm.

show abstract

Section: The Abc Algorithmmentioning

confidence: 99%

Section: The Abc Algorithmmentioning

confidence: 99%

Optimization of the Pool Boiling Heat Transfer in the Region of the Isolated Bubbles using the ABC Algorithm

Alic¹,

Çermik²,

Tokgöz³

et al. 2019

JAFM

View full text Add to dashboard Cite

show abstract

“…DNN is superior to other conventional neural networks in the classification problem with the help of the aforementioned properties by having complex decision surface. 6,7,[22][23][24][25][26][27][28][29]…”

Section: Introductionmentioning

confidence: 99%

Type 2 diabetes data classification using stacked autoencoders in deep neural networks

Kannadasan

Edla

Kuppili

2019

Clinical Epidemiology and Global Health

127

View full text Add to dashboard Cite

This paper aims to classify the Pima Indians diabetes dataset with better accuracy and other evaluation metrics. The Deep Neural Network (DNN) framework will help to diagnose the patient in an effective way with higher accuracy. Method: In this approach, we proposed a Deep Neural Network framework for diabetes data classification using stacked autoencoders. Features are extracted from the dataset using stacked autoencoders and the dataset is classified using softmax layer. Also, fine tuning of the network is done using backpropagation in supervised fashion with the training dataset. However, the medical diagnosis involves the risk factors of wrong prediction; hence we have used evaluation metrics such as precision, recall, specificity and F1-score for the evaluation of our model and have achieved better results. Results: The proposed framework is experimented on Pima Indians Diabetes data which has 768 patient records with 8 attributes for each record. We achieved classification accuracy of 86.26%. Conclusion: A stacked autoencoders based Deep Learning framework for classification of Type 2 Diabetes data is proposed in this paper. This approach is experimented on UCI machine learning data and proved the outperformance over various existing classification methods.

show abstract

“…Lavanya and Srinivasan hybridized ABC and genetic algorithm (GA) and tested mentioned hybrid optimization algorithm on solving training problem related with the neural networks [21]. Badem et al combined ABC and limited memory-based BFGS algorithms and used their new variant for training deep neural networks [22]. Badem et al also investigated the performance of their hybridized ABC algorithm on solving numerical benchmark problems [23].…”

Section: Introductionmentioning

confidence: 99%

Time-Based Dance Scheduling for Artificial Bee Colony Algorithm and Its Variants

Aslan¹

2019

IJCIS

View full text Add to dashboard Cite

Artificial Bee Colony (ABC) algorithm inspired by the intelligent source search and consumption characteristics of the real honey bees is one of the most powerful optimization techniques. Although the existing behaviours of the honey bees in standard ABC algorithm are capable of producing optimal or near optimal solutions for the vast majority of the problems, there are still some intelligent characteristics that are not modeled yet such as decision-making mechanism used by employed bees to determine when the dancing will be completed. In this study, a mechanism that adjusts the dancing durations of the employed bees according to the fitness values of the associated food sources is proposed and integrated to the standard ABC algorithm and its well-known variants. Experimental studies on a set of complex continuous numerical problems showed that the newly proposed dance scheduling approach significantly improved the search and consumption characteristics of the artificial bees of the standard ABC algorithm and its some variants.

show abstract

A new efficient training strategy for deep neural networks by hybridization of artificial bee colony and limited–memory BFGS optimization algorithms

Cited by 102 publications

References 29 publications

Optimization of the Pool Boiling Heat Transfer in the Region of the Isolated Bubbles using the ABC Algorithm

Optimization of the Pool Boiling Heat Transfer in the Region of the Isolated Bubbles using the ABC Algorithm

Type 2 diabetes data classification using stacked autoencoders in deep neural networks

Time-Based Dance Scheduling for Artificial Bee Colony Algorithm and Its Variants

Contact Info

Product

Resources

About