The Multi-Agent Behavior Dataset: Mouse Dyadic Social Interactions

Jennifer J. Sun, Tomomi Karigo, Dipam Chakraborty, Sharada Mohanty, Benjamin Wild, Quan Sun, Chen Chen, David Anderson, Pietro Perona, Yisong Yue, Ann Kennedy

Keywords: behavior modeling, trajectory data, animal behavior

TL;DR: CalMS21 is a multi-agent dataset consisting of trajectory data annotated at each frame with behavior labels from domain experts.

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.

URL: Dataset Link: https://data.caltech.edu/records/1991. Baseline code link: https://gitlab.aicrowd.com/aicrowd/research/mab-e/mab-e-baselines.

Contribution Process Agreement: Yes

Dataset Url: https://sites.google.com/view/computational-behavior/our-datasets/calms21-dataset

License: CC-BY-NC-SA

Author Statement: Yes

Community Implementations: [ 4 code implementations](https://www.catalyzex.com/paper/arxiv:2104.02710/code)