EAS CSIS Doctoral Proposal Defense by Aakash Kardam

Wednesday, November 19, 2025 at 9:00am to 11:00am

Committee chairs:

• Dr. Shelley Zhang, Department of Computer and Information Science (Co-Advisor)
• Dr. Daniel MacDonald, Department of Civil & Environmental Engineering (Co-Advisor)
• Dr. Yuchou Chang, Department of Computer and Information Science

Committee members:

• Dr. Gokhan Kul, Department of Computer and Information Science
• Dr. Eugene Chabot, Naval Undersea Warfare Center

Abstract:

Autonomous agents are increasingly deployed in mission-critical domains where adaptability, coordination, and learning are essential. This research proposes a unified framework for multi-agent learning aimed at improving team performance in dynamic, uncertain environments. The core focus lies in understanding how agents can independently learn from localized experience and, subsequently, how such knowledge can be shared and utilized across a team for collective intelligence and emergent coordination. The first branch of this research addresses single-agent learning, where agents operate independently and must adapt to varying and partially observable environmental conditions. The focus lies in enabling agents to model local dynamics, learn from sensor feedback, and autonomously adjust planning and control strategies in real time. This is especially relevant in applications like Unmanned Underwater Vehicles (UUVs), where agents must navigate challenging, data-sparse environments while maintaining mission efficiency. The second branch explores multi-agent knowledge sharing and team-level adaptation, where agents collaborate across spatially partitioned regions and exchange learned environmental insights to

improve coordination. This line of work investigates how agents can communicate partial knowledge, reassign roles dynamically, and respond to uneven environmental pressures — including the emergence of leadership behaviors, where certain agents take on coordination responsibilities without predefined roles. This is demonstrated in simulations grounded in organizational theory and formal coordination models such as TAEMS.
For further information please contact Dr. Yuchou Chang.

Dion 311
Sheryl SEARS
5089998457
ychang1@umassd.edu
https://umassd.zoom.us/j/95302822645?pwd=agFeLEMPefLLgE49E2y9hT2YrRUkW6.1