.png){kind=icon}
Return
Multiagent Inductive Policy Optimization
.png){kind=icon}
PRE
AI
DOI：10.1109/TNNLS.2025.3601360.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Policy optimization methods are promising to tackle high-complexity reinforcement learning (RL) tasks with multiple agents. In this article, we derive a general trust region for policy optimization methods by considering the effect of subpolicy combinations among agents in multiagent environments. Based on this trust region, we propose an inductive objective to train the policy function, which can ensure agents learn monotonically improving policies. Furthermore, we observe that the policy always updates very weakly before falling into a local optimum. To address this, we introduce a cost regarding policy distance in the inductive objective to strengthen the motivation of agents to explore new policies. This approach strikes a balance during training, where the policy update step size remains within the constraints of the trust region, preventing excessive updates while avoiding getting stuck in local optima. Simulations on wind farm (WF) control tasks and two multiagent benchmarks demonstrate the high performance of the proposed multiagent inductive policy optimization (MAIPO) method.
Keywords:
Training
Space exploration
Optimization methods
Q-learning
Iterative methods
Wind farms
Probability distribution
Convergence
Learning systems
Decision making
Inductive optimization objective
multiagent reinforcement learning (RL)
trust region
wind farm (WF) control
Journal
IF:
8.9
Papers:
7.6K
Citations:
7.2W
Organization
Cited Papers
ShareSaveShareSaveIntegrated adaptive communication in multi-agent systems: Dynamic topology, frequency, and content optimization for efficient collaboration.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
ShareSaveShareSave
View more