A Novel Multi-Objective Deep Q-Network: Addressing Immediate and Delayed Rewards in Multi-Objective Q-Learning

Current multi-objective reinforcement learning (MORL) research often struggles to balance multiple objectives and manage the stability and performance of learning algorithms, especially in complex environments. To address this, we propose a new multi-objective deep Q-network (MO-DQN) framework that...

Full description

Bibliographic Details

Published in:	IEEE Access
Main Author:	Youming Zhang
Format:	Article
Language:	English
Published:	IEEE 2024-01-01
Subjects:	Multi-objective reinforcement learning deep Q-network Q-learning linear scalarization
Online Access:	https://ieeexplore.ieee.org/document/10685374/