AGAVaPS - Adaptive Genetic Algorithm with Varying Population Size

Ribeiro, Rafael Rodrigues Mendes; Maciel, Carlos Dias

doi:10.1109/cec55065.2022.9870394

Cited by 2 publications

(9 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm prioritizes connection to nodes with high mutual information [19]. Finally, in [20], the adaptive genetic algorithm with varying population size (AGAVaPS) is proposed. In this algorithm, each solution has its own mutation rate and number of iterations that will be a part of the population.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bayesian Network Structural Learning Using Adaptive Genetic Algorithm with Varying Population Size

Rodrigues Mendes Ribeiro,

Dias Maciel

2023

MAKE

View full text Add to dashboard Cite

A Bayesian network (BN) is a probabilistic graphical model that can model complex and nonlinear relationships. Its structural learning from data is an NP-hard problem because of its search-space size. One method to perform structural learning is a search and score approach, which uses a search algorithm and structural score. A study comparing 15 algorithms showed that hill climbing (HC) and tabu search (TABU) performed the best overall on the tests. This work performs a deeper analysis of the application of the adaptive genetic algorithm with varying population size (AGAVaPS) on the BN structural learning problem, which a preliminary test showed that it had the potential to perform well on. AGAVaPS is a genetic algorithm that uses the concept of life, where each solution is in the population for a number of iterations. Each individual also has its own mutation rate, and there is a small probability of undergoing mutation twice. Parameter analysis of AGAVaPS in BN structural leaning was performed. Also, AGAVaPS was compared to HC and TABU for six literature datasets considering F1 score, structural Hamming distance (SHD), balanced scoring function (BSF), Bayesian information criterion (BIC), and execution time. HC and TABU performed basically the same for all the tests made. AGAVaPS performed better than the other algorithms for F1 score, SHD, and BIC, showing that it can perform well and is a good choice for BN structural learning.

show abstract

Section: Introductionmentioning

confidence: 99%

“…This algorithm was tested for different applications and performed well in problems with huge search spaces. A preliminary test on BN structural learning was also carried out and showed that it could be a good algorithm to be used on it [20].…”

Section: Introductionmentioning

confidence: 99%

Bayesian Network Structural Learning Using Adaptive Genetic Algorithm with Varying Population Size

Rodrigues Mendes Ribeiro,

Dias Maciel

2023

MAKE

View full text Add to dashboard Cite

show abstract

“…After this point, the population size reaches a kind of stability while AGAVaPS explores the search space and has not stagnated the search in some regions. Source: Ribeiro and Maciel (2022) After the stagnation happens, the population size starts to oscillate as a result of the µ mut update strategy that changes its value to the opposite size when an extreme value is reached. This results in AGAVaPS oscillating between exploitation and exploration at this point of the search.…”

Section: Agavaps Behavior Testmentioning

confidence: 99%

“…Grid Free rank 30/00/00 30/00/00 02/02/26 00/05/25 00/00/30 00/00/30 rank * 30/00/00 30/00/00 00/01/29 00/00/30 00/00/30 00/00/30 Limited rank 30/00/00 30/00/00 00/00/30 00/05/25 00/00/30 00/00/30 rank * 30/00/00 30/00/00 00/00/30 00/00/30 00/00/30 00/00/30 Both rank 60/00/00 60/00/00 02/02/56 00/10/50 00/00/60 00/00/60 rank * 60/00/00 60/00/00 00/01/59 00/00/60 00/00/60 00/00/60 Time Free rank 00/00/30 00/00/30 00/00/30 00/00/30 00/00/30 00/01/29 rank * 00/00/30 28/00/02 01/00/29 00/00/30 18/02/10 10/01/19 Limited rank 00/00/30 00/00/30 00/00/30 00/00/30 00/00/30 00/01/29 rank * 00/00/30 26/01/03 01/00/29 00/00/30 09/04/17 07/00/23 Both rank 00/00/60 00/00/60 00/00/60 00/00/60 00/00/60 00/02/58 rank * 00/00/60 54/01/05 02/00/58 00/00/60 27/06/27 17/01/42 Source: Ribeiro and Maciel (2022) Table 12 -Mean and overall rank analysis for the comparison of AGAVaPS with other algorithms for the CEC2017 benchmark. "AGAVaPS" refers to AGAVaPS with tournament size = 3 and "AGAVaPS * " refers to with AGAVaPS with tournament size = 1.…”

Section: Ga Pso Ffa Bat Csmentioning

confidence: 99%

See 1 more Smart Citation

Bayesian network quantization method and structural learning

Ribeiro

View full text Add to dashboard Cite

To my colleagues at the Signal Processing Laboratory for all the help and companionship. My thesis would not be the same without your help and suggestions during the whole process. A special thanks to Henrique and Matheus who were always at the lab and brought a lot of joy. Finally, I would like to thank Professor Maciel for helping me start my research path, from when I was an undergraduate student until now. Your help was essential for me to reach this point.

show abstract

AGAVaPS - Adaptive Genetic Algorithm with Varying Population Size

Cited by 2 publications

References 34 publications

Bayesian Network Structural Learning Using Adaptive Genetic Algorithm with Varying Population Size

Bayesian Network Structural Learning Using Adaptive Genetic Algorithm with Varying Population Size

Bayesian network quantization method and structural learning

Contact Info

Product

Resources

About