Symbiosis or Separation? Revisiting the Interface of Robot Learning and HRI

To operate effectively in human environments, robots must be able to combine actions in intelligent ways to perform complex tasks, while also communicating their intentions and capabilities to the people they interact with. Achieving this capability requires not only the ability to learn meaningful skills, but also an ability to reason about their consequences within dynamic, shared spaces.

This workshop aims to bring together researchers across multiple subfields – including computer vision, robotic skill learning, task and motion planning, and human-robot interaction (HRI) – to explore how robots can acquire and generalize useful behaviors in the context of human-centered environments. We are particularly interested in questions surrounding when and how human-subject experimentation is necessary for the development and validation of the theoretical advancement of said systems.

Why attend?

By fostering interdisciplinary dialogue, we aim to better understand the interplay between robot learning and HRI, and to explore methods for validating intelligent robotic behavior both with and without direct human involvement. This workshop will also serve as a platform to spotlight emerging research directions and to elevate the work of junior researchers pushing the boundaries of the field.

Our goal is to build new connections, surface open problems, and spark collaborations that will advance the integration of robot learning human-aware robotics in both theoretical and practical applications.

Call for Participation

Participants will have the opportunity to share their work, exchange ideas across disciplinary boundaries, and form new collaborations. This workshop is non-archival and you are welcome to submit your papers that are already accepted

Call for Submissions/Contributions

You can submit a max. 2-page extended abstract which can be accepted either as a short paper presentation or a poster. Submission will use the same format as the conference.

Submissions will be gathered using the following Google Form: https://forms.gle/iBciUeqcfHGr6Xjd8

Important Dates

Submission Deadline: June 1st, 2025 (6/1)

Notification of Acceptance: June 8th, 2025 (6/8)

Topics of interest

Robot manipulation and navigation
Learning from humans
Human-robot interaction
Physical human-robot interaction
Explainable AI
Imitation learning and reinforcement learning in robotics
Robot learning from demonstration and exploration
Human-robot trust (specifically after failures)

Tentative Schedule

12:30 - 12:35 PM | Welcome & Opening Remarks

Brief intro from organizers
Overview of workshop goals and structure

12:35 - 1:05 PM | Keynote Talk (25 min + 5 min Q&A)

TBD Topic

1:05 - 1:55 PM | Paper Presentations - Panel format (5 x 10 min)

Selected from submissions
90 second presentation on each paper then panel discussion for general Q&A
Topics spanning robot learning, HRI, or their intersection

1:55 - 2:35 PM | Coffee Break + Poster Session

Posters from remaining accepted submissions
Informal discussion and networking

2:35 - 3:35 PM | Debate & Audience Discussion

Title: “Humans: In-the-Loop or Out-of-the-Way? The Role of HRI in Robot Learning”

Debate Format: Two TBD experts, 10 min opening arguments each, 10 min rebuttals/discussion, 15-20 min moderated audience Q&A

Debate Prompts:

HRI side: “How do we convince everyone that every robotics problem is an HRI problem?”
Learning side: “How do I get rid of these pesky humans?”

3:35 - 3:50 PM | Closing Remarks & Next Steps

Recap of themes and insights
Thank participants
Invite continued discussion and potential follow-up collaborations

Keynote Speaker

TBD

Selected Contributions

Title Contribution TBD

Authors: Names and Affiliations
Abstract: Text

Title Contribution TBD

Authors: Names and Affiliations
Abstract: Text

Title Contribution TBD

Authors: Names and Affiliations
Abstract: Text

Title Contribution TBD

Authors: Names and Affiliations
Abstract: Text

Organizers