Research map / Multi-agent systems and games / Human-AI and goal-conditioned
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
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add task programming features learned from the unlabeled set (ablation) | Baseline 1D Conv Net · Task 1 | Average F1: $.793 \pm .011 -> $.829 \pm .004 | Improved |
| Add task programming features learned from the unlabeled set (ablation) | Baseline 1D Conv Net · Task 1 | MAP: $.856 \pm .010 -> $.889 \pm .004 | Improved |
| Add task programming features learned from the unlabeled set (ablation) | Baseline · Task 2 | Average F1: $.754 \pm .005 -> $.774 \pm .006 | Improved |
| Add task programming features learned from the unlabeled set (ablation) | Baseline · Task 2 | MAP: $.813 \pm .003 -> $.835 \pm .005 | Improved |
| Add task programming features learned from the unlabeled set (ablation) | Baseline · Task 3 | Average F1: 0.338 \pm .004 -> 0.328 \pm .009 | Hurt |
| Add task programming features learned from the unlabeled set (ablation) | Baseline · Task 3 | MAP: $.317 \pm .005 -> $.320 \pm .009 | Improved |
| Tune classification thresholds (ablation) | MABe 2021 Task 3 Top-1 · Task 3 | Average F1: $.319 \pm .025 -> ($.363 \pm .020) | Improved |
| Use a 1D convolutional network instead of the other evaluated architectures | Baseline sequence classifier · Task 1 | Average F1: $.610 \pm .028 -> $.793 \pm .011 | Improved |
| Use MABe 2021 Task 1 Top-1 instead of the baseline | Task 1 classifier · Task 1 | Average F1: $.793 \pm .011 -> $.864 \pm .011 | Improved |
| Use MABe 2021 Task 2 Top-1 instead of the baseline | Task 2 classifier · Task 2 | Average F1: $.754 \pm .005 -> $.809 \pm .015 | Improved |
Closest papers by meaning
- MICE: Multi-animal Interaction Context Encoder — A Hierarchical Foundation Model for Mouse Behavior (NeurIPS 2026 · similarity 0.92)
- Of Mice and Mates: Automated Classification and Modelling of Mouse Behaviour in Groups Using a Single Model Across Cages (IJCV 2024 · similarity 0.92)
- Mouse Lockbox Dataset: Behavior Recognition for Mice Solving Lockboxes (IJCV 2026 · similarity 0.90)
- MABe22: A Multi-Species Multi-Task Benchmark for Learned Representations of Behavior (ICML 2023 · similarity 0.90)
- Relax, it doesn’t matter how you get there: A new self-supervised approach for multi-timescale behavior analysis (NeurIPS 2023 · similarity 0.89)
- Agent-to-Sim: Learning Interactive Behavior Models from Casual Longitudinal Videos (ICLR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
