scholarly journals Differential Evolution for Neural Networks Optimization

Mathematics ◽  
2020 ◽  
Vol 8 (1) ◽  
pp. 69 ◽  
Author(s):  
Marco Baioletti ◽  
Gabriele Di Bari ◽  
Alfredo Milani ◽  
Valentina Poggioni
Keyword(s):  
Neural Networks ◽  
Comparative Study ◽  
Differential Evolution ◽  
Adaptive Methods ◽  
Experimental Results ◽  
Classification Problems ◽  
Adaptive Differential Evolution ◽  
Neural Networks Optimization ◽  
Better Than ◽  
Self Adaptive

In this paper, a Neural Networks optimizer based on Self-adaptive Differential Evolution is presented. This optimizer applies mutation and crossover operators in a new way, taking into account the structure of the network according to a per layer strategy. Moreover, a new crossover called interm is proposed, and a new self-adaptive version of DE called MAB-ShaDE is suggested to reduce the number of parameters. The framework has been tested on some well-known classification problems and a comparative study on the various combinations of self-adaptive methods, mutation, and crossover operators available in literature is performed. Experimental results show that DENN reaches good performances in terms of accuracy, better than or at least comparable with those obtained by backpropagation.

A Self-Adaptive Hybrid Bat Algorithm for Training Feedforward Neural Networks

2021 ◽  
Vol 12 (3) ◽  
pp. 149-171
Author(s):  
Rabab Bousmaha ◽  
Reda Mohamed Hamou ◽  
Abdelmalek Amine
Keyword(s):  
Neural Networks ◽  
Differential Evolution ◽  
Credit Card ◽  
Differential Evolution Algorithm ◽  
Solution Space ◽  
Feedforward Neural Networks ◽  
Classification Error ◽  
Training Methods ◽  
Adaptive Differential Evolution ◽  
Self Adaptive

Training feedforward neural network (FFNN) is a complex task in the supervised learning field. An FFNN trainer aims to find the best set of weights that minimizes classification error. This paper presents a new training method based on hybrid bat optimization with self-adaptive differential evolution to train the feedforward neural networks. The hybrid training algorithm combines bat and the self-adaptive differential evolution algorithm called BAT-SDE. BAT-SDE is used to better search in the solution space, which proves its effectiveness in large space solutions. The performance of the proposed approach was compared with eight evolutionary techniques and the standard momentum backpropagation and adaptive learning rate. The comparison was benchmarked and evaluated using seven bio-medical datasets and one large credit card fraud detection dataset. The results of the comparative study show that BAT-SDE outperformed other training methods in most datasets and can be an alternative to other training methods.

scholarly journals INTER AND INTRA CLUSTER ON SELF-ADAPTIVE DIFFERENTIAL EVOLUTION FOR MULTI-DOCUMENT SUMMARIZATION

2018 ◽  
Vol 11 (2) ◽  
pp. 86
Author(s):  
Alifia Puspaningrum ◽  
Adhi Nurilham ◽  
Eva Firdayanti Bisono ◽  
Khoirul Umam ◽  
Agus Zainal Arifin
Keyword(s):  
Cluster Analysis ◽  
Differential Evolution ◽  
Text Analysis ◽  
Experimental Results ◽  
Adaptive Differential Evolution ◽  
Content Coverage ◽  
Different Content ◽  
F Measure ◽  
Self Adaptive

Multi – document as one of summarization type has become more challenging issue than single-document because its larger space and its different content of each document. Hence, some of optimization algorithms consider some criteria in producing the best summary, such as relevancy, content coverage, and diversity. Those weighted criteria based on the assumption that the multi-documents are already located in the same cluster. However, in a certain condition, multi-documents consist of many categories and need to be considered too. In this paper, we propose an inter and intra cluster which consist of four weighted criteria functions (coherence, coverage, diversity, and inter-cluster analysis) to be optimized by using SaDE (Self Adaptive Differential Evolution) to get the best summary result. Therefore, the proposed method will deal not only with the value of compactness quality of the cluster within but also the separation of each cluster. Experimental results on Text Analysis Conference (TAC) 2008 datasets yields better summaries results with average ROUGE-1 on precision, recall, and f - measure 0.77, 0.07, and 0.12 compared to another method that only consider the analysis of intra-cluster.

scholarly journals Comparative performance evaluation of self-adaptive differential evolution with GA, SCE and DE algorithms for the automatic calibration of a computationally intensive distributed hydrological model

2020 ◽  
Author(s):  
Saswata Nandi ◽  
M. Janga Reddy
Keyword(s):  
Differential Evolution ◽  
Hydrological Model ◽  
Automatic Calibration ◽  
Hydrological Models ◽  
Distributed Hydrological Model ◽  
Infiltration Capacity ◽  
Vic Model ◽  
Adaptive Differential Evolution ◽  
Computationally Intensive ◽  
Self Adaptive

Abstract Recently, physically-based hydrological models have been gaining much popularity in various activities of water resources planning and management, such as assessment of basin water availability, floods, droughts, and reservoir operation. Every hydrological model contains some parameters that must be tuned to the catchment being studied to obtain reliable estimates from the model. This study evaluated the performance of different evolutionary algorithms, namely genetic algorithm (GA), shuffled complex evolution (SCE), differential evolution (DE), and self-adaptive differential evolution (SaDE) algorithm for the parameter calibration of a computationally intensive distributed hydrological model, variable infiltration capacity (VIC) model. The methodology applied and tested for a case study of the upper Tungabhadra River basin in India, and the performance of the algorithms is evaluated in terms of reliability, variability, efficacy measures in a limited number of function evaluations, their ability for achieving global convergence, and also by their capability to produce a skillful simulation of streamflows. The results of the study indicated that SaDE facilitates an effective calibration of the VIC model with higher reliability and faster convergence to optimal solutions as compared to the other methods. Moreover, due to the simplicity of the SaDE, it provides easy implementation and flexibility for the automatic calibration of complex hydrological models.

scholarly journals A face recognition framework using self-adaptive differential evolution

2019 ◽  
Vol 17 (2) ◽  
pp. 4-14
Author(s):  
Guilherme Felippe Plichoski ◽  
Chidambaram Chidambaram ◽  
Rafael Stubs Parpinelli
Keyword(s):  
Face Recognition ◽  
Differential Evolution ◽  
Adaptive Differential Evolution ◽  
Self Adaptive
Sign in / Sign up

Export Citation Format

Share Document