Call for paper: Imitation learning

Through learning from demonstration, imitation learning (IL) has become a central paradigm that enables intelligent agents to acquire complex behaviours by observing expert demonstrations. By leveraging these demonstrations, imitation learning reduces data requirements and accelerates the deployment of intelligent agents in real-world applications. This Collection welcomes research focusing on robust imitation learning under domain shift and partial observability in physical-system-driven simulators or real-world applications.

This Scientific Reports Collection welcomes original research on imitation learning. Narrative review articles are also welcomed for consideration in our sister journal Scientific Reviews. For further information, please view the ‘Participating Journals’ below.