ModGNN: Expert Policy Approximation in Multi-Agent Systems with a Modular Graph Neural Network Architecture.
Ryan KortvelesyAmanda ProrokPublished in: ICRA (2021)
Keyphrases
- multi agent systems
- inverse reinforcement learning
- graph representation
- graph theory
- error bounds
- graph structure
- agent architecture
- multi agent
- random walk
- closed form
- human experts
- intelligent agents
- bipartite graph
- agent systems
- graph model
- agent technology
- graph partitioning
- graph matching
- weighted graph
- approximation algorithms
- directed graph
- game theory
- special case
- cooperative
- autonomous agents
- optimal policy
- graph based algorithm