Agents Leveraging Learning Engagement Soldiers; .L.L..E.S

About the Research

The primary objective of the ALLIES project is to develop, test, and validate techniques for training artificial intelligence (AI) agents to work flexibly and efficiently alongside other AI agents and humans/Soldiers to accomplish tasks, goals, and the overall mission in dynamic multi‑domain operations (MDO). Specifically, this project focuses on utilizing state‑of‑the‑art multi‑agent reinforcement learning (MARL) algorithms combined with Human‑in‑the‑Loop (HitL) approaches to find a methodology for training AI agents to rapidly and flexibly adapt to, or learn from, partners (human or agent) actions to complete tasks in novel situations and environments.

DEVCOM Army Research Laboratory

ARL‑R‑CISD‑300001

multi‑agent, simulation, embodied, reinforcement learning, artificial intelligence, coordinated group actions, optimal group performance, modeling agent behavior, flexibly adaptive, human‑agent teaming, adaptive agents, behavioral analysis

Derrik Asher

ARL Advisor Email:
derrik.e.asher.civ

• Degree:
  Any degree.
• Academic Level(s):
  Any academic level.
• Discipline(s):
  • Computer, Information, and Data Sciences (Computer Science (general), Human Computer Interaction)
• Other Non‑Science & Engineering
• Age:
  Must be 18 years of age.