Skip to content

Discrepancy regarding the description of the "four domains" between Abstract and Main Text #1

@SUNDULE

Description

@SUNDULE

Dear Authors,
Thank you for your great work on the HAROOD paper. I am currently studying your research and trying to reproduce the experiments.

However, I noticed a discrepancy regarding the description of the "four domains" mentioned in the paper:

In the Abstract: In this paper, we fill this gap by proposing HAROOD, a comprehensive benchmark for HAR in OOD settings. We define 4 OOD scenarios: cross-person, cross-position, cross-dataset, and cross-time, and build a testbed covering 6 datasets, 16 comparative methods (implemented with CNN-based and Transformerbased architectures), and two model selection protocols.

In the Main Text: HAROOD includes six publicly available time-series sensor datasets and supports four domain-shift scenarios: cross-person, cross-position, cross-device, and cross-time.

Could you please clarify which definition corresponds to the actual experimental setting? I might have misunderstood some context, but I would appreciate your guidance to ensure I understand the setup correctly.

Thank you for your time!
Best regards

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions