Session: CIE-06-01: AL/ML AI and ML for System Engineering

Paper Number: 194616

194616 - Learning to Represent Social Information in Multi-Agent Reinforcement Learning for Self-Organizing Systems 

Self-organizing systems (SoS) have emerged as a promising paradigm for enabling scalable, adaptive, and resilient coordination in complex engineering tasks. Recent advances in multi-agent reinforcement learning (MARL) provide a powerful methodology for learning decentralized policies. However, exist- ing approaches lack a deep understanding of social interactions, limiting their ability to effectively utilize rich social information during learning.

This paper investigates how social learning can be system- atically integrated into MARL by focusing on the representation of social information particularly agent identity (ID). Prior stud- ies have shown that team-level behaviors often evolve into dis- tinct role specializations, suggesting that agent identity may play an important role in learning and coordination. However, how such information should be represented and incorporated into the learning process remains unclear.

To address this, we conceptualize social capability as a spec- trum and extend it by introducing agent identity as an explicit social signal. We further examine how different representa- tion methods influence learning dynamics and team-level per- formance.

Experiments are conducted on a cooperative box-pushing as- sembly task with collision avoidance, which requires complex coordination and long-term strategy formation. The results demon- strate that incorporating structured social information signifi- cantly improves learning efficiency and task performance, while revealing key mechanisms of social learning in multi-agent sys- tems. This work provides new insights into socially-aware MARL and contributes to the design of more capable self-organizing systems for complex engineering applications.

Presenting Author: Bingling Huang California State University, Fullerton

Presenting Author Biography: Education Experiences
Ph.D. in Mechanical Engineering University of Southern California (USC), Los Angeles, United States 2023
M.S. in Computer Science University of Southern California (USC), Los Angeles, United States 2021
Employment History
Assistant Professor in Dept. of Mechanical Engineering, California State University, Fullerton 08.2023 – present
Research Assistant in IMPACT Lab at University of Southern California (USC) 05.2022 – 08.2023
Machine Learning Engineer (Ph.D.) Internship in Meta, Seattle, WA 05.2021 – 08.2021