ITU

Novelty-driven Particle Swarm Optimization

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

Standard

Novelty-driven Particle Swarm Optimization. / Galvao, Diana; Lehman, Joel Anthony; Urbano, Paulo.

Artificial Evolution: 12th International Conference, Evolution Artificielle, EA 2015, Lyon, France, October 26-28, 2015. Revised Selected Papers. Springer, 2015. p. 177-190 (Lecture Notes in Computer Science, Vol. 9554).

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

Harvard

Galvao, D, Lehman, JA & Urbano, P 2015, Novelty-driven Particle Swarm Optimization. in Artificial Evolution: 12th International Conference, Evolution Artificielle, EA 2015, Lyon, France, October 26-28, 2015. Revised Selected Papers. Springer, Lecture Notes in Computer Science, vol. 9554, pp. 177-190. https://doi.org/10.1007/978-3-319-31471-6_14

APA

Galvao, D., Lehman, J. A., & Urbano, P. (2015). Novelty-driven Particle Swarm Optimization. In Artificial Evolution: 12th International Conference, Evolution Artificielle, EA 2015, Lyon, France, October 26-28, 2015. Revised Selected Papers (pp. 177-190). Springer. Lecture Notes in Computer Science Vol. 9554 https://doi.org/10.1007/978-3-319-31471-6_14

Vancouver

Galvao D, Lehman JA, Urbano P. Novelty-driven Particle Swarm Optimization. In Artificial Evolution: 12th International Conference, Evolution Artificielle, EA 2015, Lyon, France, October 26-28, 2015. Revised Selected Papers. Springer. 2015. p. 177-190. (Lecture Notes in Computer Science, Vol. 9554). https://doi.org/10.1007/978-3-319-31471-6_14

Author

Galvao, Diana ; Lehman, Joel Anthony ; Urbano, Paulo. / Novelty-driven Particle Swarm Optimization. Artificial Evolution: 12th International Conference, Evolution Artificielle, EA 2015, Lyon, France, October 26-28, 2015. Revised Selected Papers. Springer, 2015. pp. 177-190 (Lecture Notes in Computer Science, Vol. 9554).

Bibtex

@inproceedings{52dc42aedf1d4b9c930d6a6457b08d52,
title = "Novelty-driven Particle Swarm Optimization",
abstract = "Particle Swarm Optimization (PSO) is a well-known population-based optimization algorithm. Most often it is applied to optimize objective-based fitness functions that reward progress towards a desired objective or behavior. As a result, search increasingly focuses on higher-fitness areas. However, in problems with many local optima, such focus often leads to premature convergence that precludes reaching the intended objective. To remedy this problem in certain types of domains, this paper introduces Novelty-driven Particle Swarm Optimization (NdPSO), which is motivated by the novelty search algorithm in evolutionary computation. In this method particles are driven only towards instances significantly different from those found before. By ignoring the objective this way, NdPSO can circumvent the problem of deceptive local optima. Because novelty search has previously shown potential for solving tasks in genetic programming, this paper implements NdPSO as an extension of the grammatical swarm method, which combines PSO with genetic programming. The resulting NdPSO implementation is tested in three different domains representative of those in which it might provide advantage over objective-driven PSO. That is, deceptive domains in which it is easy to derive a meaningful high-level description of novel behavior. In each of the tested domains NdPSO outperforms both objective-based PSO and random-search, demonstrating its promise as a tool for solving deceptive problems.",
author = "Diana Galvao and Lehman, {Joel Anthony} and Paulo Urbano",
year = "2015",
doi = "10.1007/978-3-319-31471-6_14",
language = "English",
isbn = "978-3-319-31470-9",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "177--190",
booktitle = "Artificial Evolution",
address = "Germany",

}

RIS

TY - GEN

T1 - Novelty-driven Particle Swarm Optimization

AU - Galvao, Diana

AU - Lehman, Joel Anthony

AU - Urbano, Paulo

PY - 2015

Y1 - 2015

N2 - Particle Swarm Optimization (PSO) is a well-known population-based optimization algorithm. Most often it is applied to optimize objective-based fitness functions that reward progress towards a desired objective or behavior. As a result, search increasingly focuses on higher-fitness areas. However, in problems with many local optima, such focus often leads to premature convergence that precludes reaching the intended objective. To remedy this problem in certain types of domains, this paper introduces Novelty-driven Particle Swarm Optimization (NdPSO), which is motivated by the novelty search algorithm in evolutionary computation. In this method particles are driven only towards instances significantly different from those found before. By ignoring the objective this way, NdPSO can circumvent the problem of deceptive local optima. Because novelty search has previously shown potential for solving tasks in genetic programming, this paper implements NdPSO as an extension of the grammatical swarm method, which combines PSO with genetic programming. The resulting NdPSO implementation is tested in three different domains representative of those in which it might provide advantage over objective-driven PSO. That is, deceptive domains in which it is easy to derive a meaningful high-level description of novel behavior. In each of the tested domains NdPSO outperforms both objective-based PSO and random-search, demonstrating its promise as a tool for solving deceptive problems.

AB - Particle Swarm Optimization (PSO) is a well-known population-based optimization algorithm. Most often it is applied to optimize objective-based fitness functions that reward progress towards a desired objective or behavior. As a result, search increasingly focuses on higher-fitness areas. However, in problems with many local optima, such focus often leads to premature convergence that precludes reaching the intended objective. To remedy this problem in certain types of domains, this paper introduces Novelty-driven Particle Swarm Optimization (NdPSO), which is motivated by the novelty search algorithm in evolutionary computation. In this method particles are driven only towards instances significantly different from those found before. By ignoring the objective this way, NdPSO can circumvent the problem of deceptive local optima. Because novelty search has previously shown potential for solving tasks in genetic programming, this paper implements NdPSO as an extension of the grammatical swarm method, which combines PSO with genetic programming. The resulting NdPSO implementation is tested in three different domains representative of those in which it might provide advantage over objective-driven PSO. That is, deceptive domains in which it is easy to derive a meaningful high-level description of novel behavior. In each of the tested domains NdPSO outperforms both objective-based PSO and random-search, demonstrating its promise as a tool for solving deceptive problems.

U2 - 10.1007/978-3-319-31471-6_14

DO - 10.1007/978-3-319-31471-6_14

M3 - Article in proceedings

SN - 978-3-319-31470-9

T3 - Lecture Notes in Computer Science

SP - 177

EP - 190

BT - Artificial Evolution

PB - Springer

ER -

ID: 80993208