Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Whole-Brain Connectomic Graph Model Enables Whole-Body Locomotion Control in Fruit Fly

Zehao Jin, Yaoye Zhu, Chen Zhang, Yanan Sui

TL;DR

Fly-connectomic Graph Model is developed, whose static structure is identical to the complete connectome of an adult Drosophila for whole-body locomotion control, and demonstrates that static brain connectomes can be transformed to instantiate effective neural policy for embodied learning of movement control.

Abstract

Whole-brain biological neural networks naturally support the learning and control of whole-body movements. However, the use of brain connectomes as neural network controllers in embodied reinforcement learning remains unexplored. We investigate using the exact neural architecture of an adult fruit fly's brain for the control of its body movement. We develop Fly-connectomic Graph Model (FlyGM), whose static structure is identical to the complete connectome of an adult Drosophila for whole-body locomotion control. To perform dynamical control, FlyGM represents the static connectome as a directed message-passing graph to impose a biologically grounded information flow from sensory inputs to motor outputs. Integrated with a biomechanical fruit fly model, our method achieves stable control across diverse locomotion tasks without task-specific architectural tuning. To verify the structural advantages of the connectome-based model, we compare it against a degree-preserving rewired graph, a random graph, and multilayer perceptrons, showing that FlyGM yields higher sample efficiency and superior performance. This work demonstrates that static brain connectomes can be transformed to instantiate effective neural policy for embodied learning of movement control.

Whole-Brain Connectomic Graph Model Enables Whole-Body Locomotion Control in Fruit Fly

TL;DR

Fly-connectomic Graph Model is developed, whose static structure is identical to the complete connectome of an adult Drosophila for whole-body locomotion control, and demonstrates that static brain connectomes can be transformed to instantiate effective neural policy for embodied learning of movement control.

Abstract

Whole-brain biological neural networks naturally support the learning and control of whole-body movements. However, the use of brain connectomes as neural network controllers in embodied reinforcement learning remains unexplored. We investigate using the exact neural architecture of an adult fruit fly's brain for the control of its body movement. We develop Fly-connectomic Graph Model (FlyGM), whose static structure is identical to the complete connectome of an adult Drosophila for whole-body locomotion control. To perform dynamical control, FlyGM represents the static connectome as a directed message-passing graph to impose a biologically grounded information flow from sensory inputs to motor outputs. Integrated with a biomechanical fruit fly model, our method achieves stable control across diverse locomotion tasks without task-specific architectural tuning. To verify the structural advantages of the connectome-based model, we compare it against a degree-preserving rewired graph, a random graph, and multilayer perceptrons, showing that FlyGM yields higher sample efficiency and superior performance. This work demonstrates that static brain connectomes can be transformed to instantiate effective neural policy for embodied learning of movement control.
Paper Structure (20 sections, 11 equations, 9 figures, 1 table, 1 algorithm)

This paper contains 20 sections, 11 equations, 9 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. Connectome-structured Graph Model
  5. Training Pipeline
  6. Experiments
  7. Inductive Bias from Connectome Topology
  8. Performance on Locomotion Tasks
  9. Neural Representations Analysis
  10. Conclusion and Discussion
  11. Details of Model Architecture and Training Process
  12. Model Architecture and Implementation
  13. Training Pipeline
  14. Task-Specific Configurations
  15. Detailed Settings of Locomotion Environment
  16. ...and 5 more sections

Figures (9)

  • Figure 1: Overview of the FlyGM-enabled whole-body locomotion control framework. Observations are mapped to afferent neuron states via a lightweight input projection. Neural states are propagated through a connectome-constrained message-passing module, and the resulting efferent states are decoded into motor actions that drive the embodied Drosophila model in MuJoCo. Demonstration videos and source code is available at: https://lnsgroup.cc/research/FlyGM.
  • Figure 2: Structure of the Fly-connectomic Graph.(a) Aggregated synapse graph of the fly connectome, grouped into afferent (blue), intrinsic (green), and efferent (orange) sets across left hemibrain, central, and right hemibrain compartments. Node sizes reflect the number of neurons in each group, and arrows indicate the direction and relative strength of connectivity. (b) Force-directed graph layout Kobourov2012 of the same neural network. The spatial layout reveals hemispheric symmetry and functional clustering.
  • Figure 3: Learning efficiency and metrics across different graph topologies and architectures. (a) Total training loss; (b) mean squared error of action means ($\mu$ MSE), and (c) mean squared error of action log-standard deviations ($\sigma$ MSE) during the imitation learning stage. Shaded areas represent the standard deviation across multiple training runs. ER-Random: Erdős--Rényi random graph; Rewired: degree-preserving rewired graph; MLP: multilayer perceptron.
  • Figure 4: Gait initiation dynamics. Snapshots of the simulated fly during the gait initialization phase.
  • Figure 5: Walking dynamics. Snapshots of the simulated fly walking in a straight line at a velocity of 3 cm/s.
  • ...and 4 more figures