Artificial neural networks for motor control usually adopt generic architectures like fully connected MLPs. While general, these tabula rasa architectures rely on large amounts of experience to learn, are not easily transferable to new bodies, and have internal dynamics that are difficult to interpret. In nature, animals are born with highly structured connectivity in their nervous systems shaped by evolution; this innate circuitry acts synergistically with learning mechanisms to provide inductive biases that enable most animals to function well soon after birth and learn efficiently. Convolutional networks inspired by visual circuitry have encoded useful biases for vision. However, it is unknown the extent to which ANN architectures inspired by neural circuitry can yield useful biases for other AI domains. In this work, we ask what advantages biologically inspired ANN architecture can provide in the domain of motor control. Specifically, we translate C. elegans locomotion circuits into an ANN model controlling a simulated Swimmer agent. On a locomotion task, our architecture achieves good initial performance and asymptotic performance comparable with MLPs, while dramatically improving data efficiency and requiring orders of magnitude fewer parameters. Our architecture is interpretable and transfers to new body designs. An ablation analysis shows that constrained excitation/inhibition is crucial for learning, while weight initialization contributes to good initial performance. Our work demonstrates several advantages of biologically inspired ANN architecture and encourages future work in more complex embodied control.
From Circuit Evidence to Mechanistic Theory: An Inductive Logic Approach
Department of Computer Science, University of Manchester(曼彻斯特大学计算机科学系);Idiap Research Institute(Idiap研究机构);CRUK National Biomarker Centre, University of Manchester(曼彻斯特大学癌症研究联盟国家生物标志物中心)
The Generalist Brain Module: Module Repetition in Neural Networks in Light of the Minicolumn Hypothesis
Department of Physics, University of Oslo(奥斯陆大学物理系);Department of Numerical Analysis and Scientific Computing, Simula Research Laboratory(Simula研究实验室数值分析与科学计算系)
Cajal Neuroscience Center (CNC), CSIC(卡哈尔神经科学中心(CNC),西班牙国家科研中心);Department of Computer Science and Artificial Intelligence, University of the Basque Country (UPV/EHU)(巴斯克大学计算机科学与人工智能系)
Augmenting learning in neuro-embodied systems through neurobiological first principles
Newcastle University(纽卡斯尔大学);University of Tokyo(东京大学);University of Quebec in Trois-Rivieres(魁北克大学三河分校);IT:U Interdisciplinary Transformation University Austria(奥地利跨学科转型大学);Okinawa Institute of Science and Technology Graduate University(冲绳科学技术大学院大学)
Quantum HyperNetworks: Training Binary Neural Networks in Quantum Superposition
Institute for Theoretical Physics, ETH Zürich(瑞士苏黎世联邦理工学院理论物理研究所);Vector Institute, MaRS Centre(向量研究所马尔中心);Department of Physics and Astronomy, University of Waterloo(滑铁卢大学物理与天文学系);Perimeter Institute for Theoretical Physics, Waterloo(滑铁卢大学理论物理研究所);University of Toronto(多伦多大学);School of Engineering, University of Guelph(圭尔夫大学工程学院);AWS Center for Quantum Computing, Pasadena, CA(亚当斯韦斯研究所量子计算中心)
Posterior-Calibrated Causal Circuits in Variational Autoencoders: Why Image-Domain Interpretability Fails on Tabular Data
Indian Institute of Technology, Patna(印度理工学院巴特那分校);Rajarshi School of Management Technology, Varanasi(拉贾尔希管理技术学院瓦拉纳西分校);Jaypee Institute of Information Technology, Noida(杰皮信息技术学院诺伊达分校)