A-SOiD, an active learning platform for expert-guided, data efficient discovery of behavior

preprint OA: closed
📄 Open PDF View at publisher

Abstract

To identify and extract naturalistic behavior, two schools of methods have become popular: supervised and unsupervised. Each approach carries its own strengths and weaknesses, which the user must weigh in on their decision. Here, a new active learning platform, A-SOiD, blends these strengths and, in doing so, overcomes several of their inherent drawbacks. A-SOiD iteratively learns user-defined groups and can considerably reduce the necessary training data while attaining expansive classification through directed unsupervised classification. In socially-interacting mice, A-SOiD outperformed other methods and required 85% less training data than was available. Additionally, it isolated two additional ethologically-distinct mouse interactions via unsupervised classification. Similar performance and efficiency were observed using non-human primate 3D pose data. In both cases, the transparency in A-SOiD's cluster definitions revealed the defining features of the supervised classification through a game-theoretic approach. Lastly, we show the potential of A-SOiD to segment a large and rich variety of human social and single-person behaviors with 3D position keypoints. To facilitate use, A-SOiD comes as an intuitive, open-source interface for efficient segmentation of user-defined behaviors and discovered subactions.

My notes (saved in your browser only)

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. The paper's references may be in our DB but unresolved to ``paper_id`` (resolution happens at ingest when the cited DOI matches a row we already have). Run the cross-source citation reconcile pass to retry.

Source provenance

europepmc
last seen: 2026-05-19T01:45:01.086888+00:00