Title | ||
---|---|---|
Moving away from error-based learning in multi-objective estimation of distribution algorithms |
Abstract | ||
---|---|---|
In this work we analyze the model-building issue and the requirements it imposes on the learning paradigm being used. We argue that error-based learning, the class of learning most commonly used in MOEDAs, is responsible for current MOEDA underachievement. We present ART as a viable alternative and present a novel algorithm called multi-objective ART-based EDA (MARTEDA) that uses a Gaussian ART neural network for model-building and an hypervolume based selector as described for the HypE algorithm. We experimentally show that thanks to MARTEDA's novel model-building approach and an indicator-based population ranking the algorithm it is able to outperform similar MOEDAs and MOEAs. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1145/1830483.1830585 | GECCO |
Keywords | Field | DocType |
model-building issue,gaussian art neural network,error-based learning,multi-objective estimation,hype algorithm,indicator-based population,present art,multi-objective art-based eda,current moeda underachievement,similar moedas,novel algorithm,distribution algorithm,adaptive resonance theory,estimation of distribution algorithm,neural network,estimation of distribution algorithms,multi objective optimization,model building | Adaptive resonance theory,Population,Mathematical optimization,Estimation of distribution algorithm,Ranking,Computer science,Multi-objective optimization,Gaussian,Artificial intelligence,Artificial neural network,Population-based incremental learning,Machine learning | Conference |
Citations | PageRank | References |
1 | 0.35 | 3 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Luis Martí | 1 | 100 | 7.73 |
Jesús García | 2 | 31 | 1.97 |
Antonio Berlanga | 3 | 1 | 0.35 |
José M. Molina | 4 | 604 | 67.82 |