A framework for controllable Pareto front learning with completed scalarization functions and its applications.

Autor: Tuan TA; School of Applied Mathematics and Informatics, Hanoi University of Science and Technology, Ha Noi, Viet Nam. Electronic address: tuan.ta181295@sis.hust.edu.vn., Hoang LP; College of Engineering and Computer Science, VinUniversity, Ha Noi, Viet Nam. Electronic address: long.hp@vinuni.edu.vn., Le DD; College of Engineering and Computer Science, VinUniversity, Ha Noi, Viet Nam. Electronic address: dung.ld@vinuni.edu.vn., Thang TN; School of Applied Mathematics and Informatics, Hanoi University of Science and Technology, Ha Noi, Viet Nam. Electronic address: thang.tranngoc@hust.edu.vn.
Jazyk: angličtina
Zdroj: Neural networks : the official journal of the International Neural Network Society [Neural Netw] 2024 Jan; Vol. 169, pp. 257-273. Date of Electronic Publication: 2023 Oct 28.
DOI: 10.1016/j.neunet.2023.10.029
Abstrakt: Pareto Front Learning (PFL) was recently introduced as an efficient method for approximating the entire Pareto front, the set of all optimal solutions to a Multi-Objective Optimization (MOO) problem. In the previous work, the mapping between a preference vector and a Pareto optimal solution is still ambiguous, rendering its results. This study demonstrates the convergence and completion aspects of solving MOO with pseudoconvex scalarization functions and combines them into Hypernetwork in order to offer a comprehensive framework for PFL, called Controllable Pareto Front Learning. Extensive experiments demonstrate that our approach is highly accurate and significantly less computationally expensive than prior methods in term of inference time.
Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Tran Ngoc Thang reports financial support was provided by Viet Nam Ministry of Education and Training.
(Copyright © 2023 Elsevier Ltd. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: