跳到正文
arXiv cs.AI· Nikhil X. Bhattasali, Jaron Cui, Lerrel Pinto, Grace W. Lindsay·· 10 小时前AI 评分31

Rat NCAP:基于哺乳动物脊髓回路的具身大鼠运动神经架构

Neural Circuit Architectural Priors for Rat Locomotion

AI 导读

研究者提出 Rat v2 生物力学大鼠身体与 Rat NCAP 神经架构,后者基于哺乳动物脊髓回路,引入细胞类型特异性连接和新型屈伸肌神经机械接口。模型无需模仿学习即可学会不同速度运动与步态切换,其运动统计、运动学和细胞类型特异性神经活动与实验数据一致。GPU 加速仿真训练基础设施支撑了电路参数优化,硅内细胞类型消融复现已知步态表型并预测渐变扰动下的步态变化。

正文 · 原文

View PDF HTML (experimental)

Abstract:Animals generate behavior through continuous interactions between their neural circuits, bodies, and environment. Embodied simulations study these interactions by testing neural circuit models in biologically realistic settings. Recent efforts have simulated diverse animals, and emerging work on biologically grounded neural architectures has begun to integrate circuit connectivity and single-unit dynamics from experimental data into artificial neural networks, introducing stronger inductive biases and closer mechanistic alignment with biology. How can we develop embodied simulations that enable the study of increasingly complex bodies and neural circuits? In this work, we address the challenge through coordinated advances in the simulated body, neural architecture, and training infrastructure. We introduce Rat v2, an efficient and flexible biomechanical body for simulating rat behavior. To control this body, we develop Rat NCAP, a biologically grounded neural architecture for locomotion based on mammalian spinal circuits that incorporates cell-type-specific connectivity and a novel flexor-extensor neuromechanical interface. To optimize circuit parameters, we develop GPU-accelerated simulation and training infrastructure. The model learns to locomote at different speeds and transition between gaits without imitation learning, producing locomotion statistics, kinematics, and cell-type-specific neural activity consistent with experimental data. Analyses reveal that the neuromechanical interface substantially affects trainability and gait quality, and in silico cell-type ablations reproduce known gait phenotypes and predict gait changes under graded perturbations. Together, these findings support the view that embodied simulations at an intermediate level of abstraction can simultaneously reproduce complex behavior and yield mechanistic insights into the neural circuits that generate it.
Comments: NeurIPS 2026. 10 pages main text, 41 pages total, 7 figures in main text. Code and videos: this https URL
Subjects: Neurons and Cognition (q-bio.NC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Robotics (cs.RO)
Cite as: arXiv:2611.23364 [q-bio.NC]
  (or arXiv:2611.23364v1 [q-bio.NC] for this version)
  https://doi.org/10.48550/arXiv.2611.23364

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Nikhil Bhattasali [view email]
[v1] Tue, 17 Nov 2026 18:57:31 UTC (7,517 KB)

来源:arXiv cs.AI · arxiv.org