Biasing the transition of Bayesian optimization algorithm between Markov chain states in dynamic environments
| Autor: | Marjan Kaedi, Chang Wook Ahn, Nasser Ghasem-Aghaee |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Mathematical optimization
Information Systems and Management Stationary distribution Markov chain Computer science 05 social sciences Evolutionary algorithm 050301 education Bayesian network Markov chain Monte Carlo 02 engineering and technology Variable-order Bayesian network Causal Markov condition Computer Science Applications Theoretical Computer Science symbols.namesake Metropolis–Hastings algorithm Artificial Intelligence Control and Systems Engineering 0202 electrical engineering electronic engineering information engineering symbols 020201 artificial intelligence & image processing 0503 education Software |
| Zdroj: | Information Sciences. :44-64 |
| ISSN: | 0020-0255 |
| DOI: | 10.1016/j.ins.2015.11.030 |
| Popis: | When memory-based evolutionary algorithms are applied in dynamic environments, the certainly use of uncertain prior knowledge for future environments may mislead the evolutionary algorithms. To address this problem, this paper presents a new, memory-based evolutionary approach for applying the Bayesian optimization algorithm (BOA) in dynamic environments. Our proposed method, unlike existing memory-based methods, uses the knowledge of former environments probabilistically in future environments. For this purpose, the run of BOA is modeled as the movements in a Markov chain, in which the states become the Bayesian networks that are learned in every generation. When the environment changes, a stationary distribution of the Markov chain is defined on the basis of the retrieved prior knowledge. Then, the transition probabilities of BOA in the Markov chain are modified (biased) to comply with the defined stationary distribution. To this end, we employ the Metropolis algorithm and modify the K2 algorithm for learning the Bayesian network in BOA in order to reflect the obtained transition probabilities. Experimental results show that the proposed method achieves improved performance compared to conventional methods, especially in random environments. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect