WIP: Human-AI interactions in real-world complex environments using a comprehensive reinforcement learning framework | ALA Workshop @ AAMAS 2023

This collaborative work was presented at the Adaptive and Learning Agents (ALA) Workshop at the 22nd International Conference on Autonomous Agents and Multiagent Systems in May 2023 (AAMAS).

Further information about AIR’s presence at AAMAS 2023 are available in this blog post.

Abstract

Deep reinforcement learning (RL) has successfully tackled many real-world tasks. However, these algorithms suffer from the well- known sample-inefficiency problem. Deep RL systems usually re- quire millions of environment interactions to learn and have stable performance. In this work, we show that human-AI teams out- perform human-only controlled and fully autonomous teams for complex tasks. We develop a novel simulator for a critical infrastruc- ture scenario and a user interface for humans to effectively advise AI agents. We show that humans can provide useful advice to the RL agents, allowing them to improve learning in a multi-agent setting.

Cite

@inproceedings{humanaiinteractions2023,
    title={
        {WIP}: Human-AI interactions in real-world complex environments using a comprehensive reinforcement learning framework
    }, 
      author={
        Islam, Md Saiful and 
        Das, Srijita and 
        Gottipati, Sai Krishna and 
        Duguay, William and 
        Mars, Clodéric and 
        Arabneydi, Jalal and 
        Fagette, Antoine and 
        Guzdial, Matthew and 
        Taylor, Matthew E.
    },
    booktitle={Adaptive Learning Agents Workshop, ALA 2023, Held as Part of the AAMAS 2023}, 
    year={2023}
}