Aerospace Controls Laboratory

Amazon Research Awards-Multiagent Reinforcement Learning

There is a critical need to develop versatile artificial intelligence (AI) agents capable of solving various complex missions. However, conventional AI systems based on centralized learning are difficult to scale up: they have limitations of the high cost of maintaining big data and large models, the inefficiency of learning each different task from scratch, and lack of reliability due to central node failures. To address these issues, we have developed various multiagent reinforcement learning frameworks, in which distributed AI agents share pertinent knowledge, learn generalized joint policies across related tasks, and coordinate with each other to achieve task objectives efficiently. Recently, we are exploring a new multiagent reinforcement learning framework based on meta-learning to enable agents to adapt fast with respect to the fellow agents’ non-stationary policies.

Related Publications

  • Wadhwania, S., Kim, D.-K., Omidshafiei, S., and How, J. P., “Policy Distillation and Value Matching in Multiagent Reinforcement Learning,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Macau, China: 2019.
  • Lutjens, B., Everett, M., and How, J. P., “Certified Adversarial Robustness for Deep Reinforcement Learning,” Conference on Robot Learning (CoRL), Osaka, Japan: 2019.
  • Downes, L. M., Steiner, T. J., and How, J. P., “Lunar Terrain Relative Navigation Using a Convolutional Neural Network for Visual Crater Detection,” American Control Conference, 2020.