MIMIC-MJX: Neuromechanical Emulation of Animal Behavior
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908686232846336 |
|---|---|
| author | Zhang, Charles Y. Yang, Yuanjia Sirbu, Aidan Abe, Elliott T. T. Wärnberg, Emil Leonardis, Eric J. Aldarondo, Diego E. Lee, Adam Prasad, Aaditya Foat, Jason Bian, Kaiwen Park, Joshua Bhatt, Rusham Saunders, Hutton Nagamori, Akira Thanawalla, Ayesha R. Huang, Kee Wui Plum, Fabian Beck, Hendrik K. Flavell, Steven W. Labonte, David Richards, Blake A. Brunton, Bingni W. Azim, Eiman Ölveczky, Bence P. Pereira, Talmo D. |
| author_facet | Zhang, Charles Y. Yang, Yuanjia Sirbu, Aidan Abe, Elliott T. T. Wärnberg, Emil Leonardis, Eric J. Aldarondo, Diego E. Lee, Adam Prasad, Aaditya Foat, Jason Bian, Kaiwen Park, Joshua Bhatt, Rusham Saunders, Hutton Nagamori, Akira Thanawalla, Ayesha R. Huang, Kee Wui Plum, Fabian Beck, Hendrik K. Flavell, Steven W. Labonte, David Richards, Blake A. Brunton, Bingni W. Azim, Eiman Ölveczky, Bence P. Pereira, Talmo D. |
| contents | The primary output of the nervous system is movement and behavior. While recent advances have democratized pose tracking during complex behavior, kinematic trajectories alone provide only indirect access to the underlying control processes. Here we present MIMIC-MJX, a framework for learning biologically-plausible neural control policies from kinematics. MIMIC-MJX models the generative process of motor control by training neural controllers that learn to actuate biomechanically-realistic body models in physics simulation to reproduce real kinematic trajectories. We demonstrate that our implementation is accurate, fast, data-efficient, and generalizable to diverse animal body models. Policies trained with MIMIC-MJX can be utilized to both analyze neural control strategies and simulate behavioral experiments, illustrating its potential as an integrative modeling framework for neuroscience. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_20532 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MIMIC-MJX: Neuromechanical Emulation of Animal Behavior Zhang, Charles Y. Yang, Yuanjia Sirbu, Aidan Abe, Elliott T. T. Wärnberg, Emil Leonardis, Eric J. Aldarondo, Diego E. Lee, Adam Prasad, Aaditya Foat, Jason Bian, Kaiwen Park, Joshua Bhatt, Rusham Saunders, Hutton Nagamori, Akira Thanawalla, Ayesha R. Huang, Kee Wui Plum, Fabian Beck, Hendrik K. Flavell, Steven W. Labonte, David Richards, Blake A. Brunton, Bingni W. Azim, Eiman Ölveczky, Bence P. Pereira, Talmo D. Neurons and Cognition Artificial Intelligence Robotics The primary output of the nervous system is movement and behavior. While recent advances have democratized pose tracking during complex behavior, kinematic trajectories alone provide only indirect access to the underlying control processes. Here we present MIMIC-MJX, a framework for learning biologically-plausible neural control policies from kinematics. MIMIC-MJX models the generative process of motor control by training neural controllers that learn to actuate biomechanically-realistic body models in physics simulation to reproduce real kinematic trajectories. We demonstrate that our implementation is accurate, fast, data-efficient, and generalizable to diverse animal body models. Policies trained with MIMIC-MJX can be utilized to both analyze neural control strategies and simulate behavioral experiments, illustrating its potential as an integrative modeling framework for neuroscience. |
| title | MIMIC-MJX: Neuromechanical Emulation of Animal Behavior |
| topic | Neurons and Cognition Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2511.20532 |