Michael Everett

R-MADDPG for Partially Observable Environments and Limited Communication

2020 · arXiv (Cornell University)

There are several real-world tasks that would benefit from applying multiagent reinforcement learning (MARL) algorithms, including the coordination among self-driving cars. The real world has challenging conditions for multiagent learning systems, such as its partial …

Neural Network Verification in Control

2021 · 2021 60th IEEE Conference on Decision and Control (CDC)

Learning-based methods could provide solutions to many of the long-standing challenges in control. However, the neural networks (NNs) commonly used in modern learning approaches present substantial challenges for analyzing the resulting control systems’ safety properties. …