Survey on Graph Neural Networks for Imitation Learning: Algorithms and Applications

Abstract

This survey examines the integration of graph neural networks (GNNs) and imitation learning (IL), focusing on the algorithms that merge these technologies and their practical applications. It explores how GNNs are used to effectively embed task-relevant information and how IL enables agents to replicate complex tasks. The review covers several key algorithms and examines their use in robotics, transportation, and engineering, highlighting the potential of GNNs and IL to enhance decision-making and performance in intricate environments.