Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent
V. Steinhoff, P. Kerschke, P. Aspar, H. Trautmann, C. Grimme, in: Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, Australia, 2020, pp. 2445–2452.
Download
No fulltext has been uploaded.
Conference Paper
| English
Author
Steinhoff, Vera;
Kerschke, Pascal;
Aspar, Pelin;
Trautmann, HeikeLibreCat 
;
Grimme, Christian
;
Grimme, ChristianAbstract
Multimodality is one of the biggest difficulties for optimization as local optima are often preventing algorithms from making progress. This does not only challenge local strategies that can get stuck. It also hinders meta-heuristics like evolutionary algorithms in convergence to the global optimum. In this paper we present a new concept of gradient descent, which is able to escape local traps. It relies on multiobjectivization of the original problem and applies the recently proposed and here slightly modified multi-objective local search mechanism MOGSA. We use a sophisticated visualization technique for multi-objective problems to prove the working principle of our idea. As such, this work highlights the transfer of new insights from the multi-objective to the single-objective domain and provides first visual evidence that multiobjectivization can link single-objective local optima in multimodal landscapes.
Publishing Year
Proceedings Title
Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)
Page
2445–2452
LibreCat-ID
Cite this
Steinhoff V, Kerschke P, Aspar P, Trautmann H, Grimme C. Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent. In: Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI). ; 2020:2445–2452. doi:10.1109/SSCI47803.2020.9308259
Steinhoff, V., Kerschke, P., Aspar, P., Trautmann, H., & Grimme, C. (2020). Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent. Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), 2445–2452. https://doi.org/10.1109/SSCI47803.2020.9308259
@inproceedings{Steinhoff_Kerschke_Aspar_Trautmann_Grimme_2020, place={Canberra, Australia}, title={Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent}, DOI={10.1109/SSCI47803.2020.9308259}, booktitle={Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)}, author={Steinhoff, Vera and Kerschke, Pascal and Aspar, Pelin and Trautmann, Heike and Grimme, Christian}, year={2020}, pages={2445–2452} }
Steinhoff, Vera, Pascal Kerschke, Pelin Aspar, Heike Trautmann, and Christian Grimme. “Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent.” In Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), 2445–2452. Canberra, Australia, 2020. https://doi.org/10.1109/SSCI47803.2020.9308259.
V. Steinhoff, P. Kerschke, P. Aspar, H. Trautmann, and C. Grimme, “Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent,” in Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), 2020, pp. 2445–2452, doi: 10.1109/SSCI47803.2020.9308259.
Steinhoff, Vera, et al. “Multiobjectivization of Local Search: Single-Objective Optimization Benefits From Multi-Objective Gradient Descent.” Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), 2020, pp. 2445–2452, doi:10.1109/SSCI47803.2020.9308259.
Google Scholar