The Multi-Agent Behavior Dataset: Mouse Dyadic Social Interactions

TL;DR

This paper introduces the CalMS21 dataset, a comprehensive collection of mouse social interaction videos with annotations, designed to facilitate automated behavior classification and study inter-annotator differences in behavioral neuroscience.

Contribution

The CalMS21 dataset provides a large-scale, annotated dataset for multi-agent behavior modeling, including benchmarks for behavior classification, style transfer, and learning with limited data.

Findings

6 million frames of tracked mouse poses

Over 1 million frames with behavior annotations

Benchmarks for behavior classification and style transfer

Abstract

Multi-agent behavior modeling aims to understand the interactions that occur between agents. We present a multi-agent dataset from behavioral neuroscience, the Caltech Mouse Social Interactions (CalMS21) Dataset. Our dataset consists of trajectory data of social interactions, recorded from videos of freely behaving mice in a standard resident-intruder assay. To help accelerate behavioral studies, the CalMS21 dataset provides benchmarks to evaluate the performance of automated behavior classification methods in three settings: (1) for training on large behavioral datasets all annotated by a single annotator, (2) for style transfer to learn inter-annotator differences in behavior definitions, and (3) for learning of new behaviors of interest given limited training data. The dataset consists of 6 million frames of unlabeled tracked poses of interacting mice, as well as over 1 million frames with tracked poses and corresponding frame-level behavior annotations. The challenge of our dataset is to be able to classify behaviors accurately using both labeled and unlabeled tracking data, as well as being able to generalize to new settings.