A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network
Particle swarm optimisation algorithm (PSO) possesses a strong exploitation capability due to its fast search speed. It, however, suffers from an early convergence leading to its inability to preserve diversity. An improved particle swarm optimiser is proposed based on a constriction factor and Grav...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2022-12-01
|
| Series: | Connection Science |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/09540091.2021.2025339 |
| _version_ | 1797684034664398848 |
|---|---|
| author | Jeremiah Osei-kwakye Fei Han Alfred Adutwum Amponsah Qinghua Ling Timothy Apasiba Abeo |
| author_facet | Jeremiah Osei-kwakye Fei Han Alfred Adutwum Amponsah Qinghua Ling Timothy Apasiba Abeo |
| author_sort | Jeremiah Osei-kwakye |
| collection | DOAJ |
| description | Particle swarm optimisation algorithm (PSO) possesses a strong exploitation capability due to its fast search speed. It, however, suffers from an early convergence leading to its inability to preserve diversity. An improved particle swarm optimiser is proposed based on a constriction factor and Gravitational Search Algorithm to overcome premature convergence. The constriction factor ensures an appropriately controlled transition from exploration into exploitation, leading to an enhanced diversity and appropriate learning rate adjustment throughout the search process. We introduce Gravitational Search Algorithm to enhance the exploratory ability of PSO. An adaptive response strategy is incorporated to activate stagnated particles to curtail the high tendency to get trapped in a local optimum. To verify the efficacy of the improvement strategies, we employ the proposed algorithm in training a Single Layer Feedforward neural network to classify real-world data ranging from binary to multi-class datasets of which our proposed algorithm outperforms the others. |
| first_indexed | 2024-03-12T00:24:42Z |
| format | Article |
| id | doaj.art-dfe3ef4796354e43b2c452cda79c4f15 |
| institution | Directory Open Access Journal |
| issn | 0954-0091 1360-0494 |
| language | English |
| last_indexed | 2024-03-12T00:24:42Z |
| publishDate | 2022-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Connection Science |
| spelling | doaj.art-dfe3ef4796354e43b2c452cda79c4f152023-09-15T10:48:00ZengTaylor & Francis GroupConnection Science0954-00911360-04942022-12-0134157860710.1080/09540091.2021.20253392025339A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural networkJeremiah Osei-kwakye0Fei Han1Alfred Adutwum Amponsah2Qinghua Ling3Timothy Apasiba Abeo4Jiangsu UniversityJiangsu UniversityJiangsu UniversityJiangsu University of Science and TechnologyTamale Technical UniversityParticle swarm optimisation algorithm (PSO) possesses a strong exploitation capability due to its fast search speed. It, however, suffers from an early convergence leading to its inability to preserve diversity. An improved particle swarm optimiser is proposed based on a constriction factor and Gravitational Search Algorithm to overcome premature convergence. The constriction factor ensures an appropriately controlled transition from exploration into exploitation, leading to an enhanced diversity and appropriate learning rate adjustment throughout the search process. We introduce Gravitational Search Algorithm to enhance the exploratory ability of PSO. An adaptive response strategy is incorporated to activate stagnated particles to curtail the high tendency to get trapped in a local optimum. To verify the efficacy of the improvement strategies, we employ the proposed algorithm in training a Single Layer Feedforward neural network to classify real-world data ranging from binary to multi-class datasets of which our proposed algorithm outperforms the others.http://dx.doi.org/10.1080/09540091.2021.2025339gravitational search algorithmparticle swarm optimisationadaptive response strategyfeedforward neural network |
| spellingShingle | Jeremiah Osei-kwakye Fei Han Alfred Adutwum Amponsah Qinghua Ling Timothy Apasiba Abeo A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network Connection Science gravitational search algorithm particle swarm optimisation adaptive response strategy feedforward neural network |
| title | A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network |
| title_full | A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network |
| title_fullStr | A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network |
| title_full_unstemmed | A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network |
| title_short | A hybrid optimization method by incorporating adaptive response strategy for Feedforward neural network |
| title_sort | hybrid optimization method by incorporating adaptive response strategy for feedforward neural network |
| topic | gravitational search algorithm particle swarm optimisation adaptive response strategy feedforward neural network |
| url | http://dx.doi.org/10.1080/09540091.2021.2025339 |
| work_keys_str_mv | AT jeremiahoseikwakye ahybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT feihan ahybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT alfredadutwumamponsah ahybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT qinghualing ahybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT timothyapasibaabeo ahybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT jeremiahoseikwakye hybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT feihan hybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT alfredadutwumamponsah hybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT qinghualing hybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork AT timothyapasibaabeo hybridoptimizationmethodbyincorporatingadaptiveresponsestrategyforfeedforwardneuralnetwork |