Publications

• Sorted by Date • Classified by Publication Type • Classified by Research Category •

SHARPIE: A Modular Framework for Reinforcement Learning and Human-AI Interaction Experiments

Libio Goncalves Braz, Kevin Godin-Dubois, Floris Den Hengst, Hüseyin Aydın, Kim Baraka, Frans A Oliehoek, and Shihan Wang. SHARPIE: A Modular Framework for Reinforcement Learning and Human-AI Interaction Experiments. In European Workshop on Reinforcement Learning (EWRL), October 2026.

Download

pdf [421.3kB]  

Abstract

Reinforcement Learning (RL) offers a general approach for modeling and training AI agents, potentially including human-AI interaction scenarios. In this paper, we propose SHARPIE (Shared Human-AI Reinforcement Learning Platform for Interactive Experiments) to address the need for a generic framework to support experiments with RL agents and humans. Its modular design consists of a versatile wrapper for RL environments and algorithm libraries, a participant-facing web interface, logging utilities, deployment on popular cloud and participant recruitment platforms. It empowers researchers to study a wide variety of research questions related to the interaction between humans and RL agents, including those related to interactive reward specification and learning, learning from human feedback, action delegation, preference elicitation, user-modeling, and human-AI teaming. The platform is based on a generic interface for human-RL interactions that aims to standardize the field of study on RL in human contexts. To demonstrate ease-of-use and versatility, we replicate four illustrative experiments from the multi-agent RL literature and report on the platform's performance under stress testing to showcase its robustness.

BibTeX Entry

@inproceedings{Braz26EWRL,
    title =     {{SHARPIE}: A Modular Framework for Reinforcement Learning and Human-AI Interaction Experiments},
    author =    {{Goncalves Braz}, Libio
                 and Kevin Godin-Dubois 
                 and {Den Hengst}, Floris 
                 and Ayd{\i}n, H{\"u}seyin 
                 and Kim Baraka
                 and Frans A Oliehoek
                 and Shihan Wang},
    booktitle=  {European Workshop on Reinforcement Learning (EWRL)},
    year=       2026,
    month =     oct,
    OPTurl=        {https://openreview.net/forum?id=NnVnCEPt55},
    keywords =   {refereed},
    abstract =  {
    Reinforcement Learning (RL) offers a general approach for modeling and training
    AI agents, potentially including human-AI interaction scenarios. In this
    paper, we propose SHARPIE (Shared Human-AI Reinforcement Learning Platform
    for Interactive Experiments) to address the need for a generic
    framework to support experiments with RL agents and humans. Its modular
    design consists of a versatile wrapper for RL environments and algorithm
    libraries, a participant-facing web interface, logging utilities,
    deployment on popular cloud and participant recruitment platforms. It empowers
    researchers to study a wide variety of research questions related to the
    interaction between humans and RL agents, including those related to
    interactive reward specification and learning, learning from human
    feedback, action delegation, preference elicitation, user-modeling, and
    human-AI teaming. The platform is based on a generic interface for human-RL
    interactions that aims to standardize the field of study on RL in human
    contexts. To demonstrate ease-of-use and versatility, we replicate four
    illustrative experiments from the multi-agent RL literature and report on
    the platform's performance under stress testing to showcase its robustness.
    }
}

Generated by bib2html.pl (written by Patrick Riley) on Thu Oct 08, 2026 13:51:42 UTC