Deep Reinforcement Learning for Multiobjective Optimization

• Published in: IEEE transactions on cybernetics Vol. 51; no. 6; pp. 3103 - 3114
• Main Authors: Li, Kaiwen, Zhang, Tao, Wang, Rui
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.06.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Decomposition; Deep learning; Deep reinforcement learning (DRL); Iterative methods; Machine learning; Mathematical models; Modeling; Mopping; multiobjective optimization; Multiple objective analysis; Neural networks; Optimization; Parameters; Pareto optimization; Pointer Network; Reinforcement learning; Training; Traveling salesman problem; Traveling salesman problems; Urban areas

This article proposes an end-to-end framework for solving multiobjective optimization problems (MOPs) using deep reinforcement learning (DRL), that we call DRL-based multiobjective optimization algorithm (DRL-MOA). The idea of decomposition is adopted to decompose the MOP into a set of scalar optimi...