Convolutional neural networks (CNNs) have recently emerged as promising models of the ventral visual stream, despite their lack of biological specificity. While current state-of-the-art models of the primary visual cortex (V1) have surfaced from training with adversarial examples and extensively augmented data, these models are still unable to explain key neural properties observed in V1 that arise from biological circuitry. To address this gap, we systematically incorporated neuroscience-derived architectural components into CNNs to identify a set of mechanisms and architectures that comprehensively explain neural activity in V1. We show drastic improvements in model-V1 alignment driven by the integration of architectural components that simulate center-surround antagonism, local receptive fields, tuned normalization, and cortical magnification. Upon enhancing task-driven CNNs with a collection of these specialized components, we uncover models with latent representations that yield state-of-the-art explanation of V1 neural activity and tuning properties. Our results highlight an important advancement in the field of NeuroAI, as we systematically establish a set of architectural components that contribute to unprecedented explanation of V1. The neuroscience insights that could be gleaned from increasingly accurate in-silico models of the brain have the potential to greatly advance the fields of both neuroscience and artificial intelligence.
Harnessing cortical geometry, wiring, and function as inductive biases for recurrent neural networks
Neuromatch Academy, Neuromatch, Inc., USA(Neuromatch学院,Neuromatch公司,美国);McGovern Institute for Brain Research, Massachusetts Institute of Technology (MIT)(麦戈文脑科学研究所,麻省理工学院(MIT))
Beyond accuracy: generalization properties of bio-plausible temporal credit assignment rules
University of Washington(华盛顿大学);Allen Institute for Brain Science(艾伦脑科学研究所);McGill University(麦吉尔大学);Mila - Quebec AI Institute(米拉-魁北克人工智能研究所);Montreal Neurological Institute(蒙特利尔神经学研究所);Université de Montréal(蒙特利尔大学)
A biologically constrained model of the whole basal ganglia addressing the paradoxes of connections and selection
Institut des Systèmes Intelligents et de Robotique(智能系统与机器人研究所);Université Pierre et Marie CURIE(皮埃尔和玛丽·居里大学);UPMC Univ Paris 06(巴黎第六大学(皮埃尔和玛丽·居里大学));CNRS(法国国家科学研究中心)
Fine-Pruning: A Biologically Inspired Algorithm for Personalization of Machine Learning Models
College of Computing, The Georgia Institute of Technology(佐治亚理工学院计算学院);Taub Faculty of Computer Science, Technion - Israel Institute of Technology(以色列理工学院陶布计算机科学学院);Faculty of Biology, Technion - Israel Institute of Technology(以色列理工学院生物学院)
Discovering Subtypes of Neurodegenerative Progression with a Scalable Connectome-Constrained Dynamic Model
Illinois Institute of Technology(伊利诺伊理工学院);Amsterdam UMC(阿姆斯特丹大学医学中心);Inria, Université Côte d’Azur(法国国家信息与自动化研究所,蔚蓝海岸大学);University of Southern California(南加州大学)
NOBLE -- Neural Operator with Biologically-informed Latent Embeddings to Capture Experimental Variability in Biological Neuron Models
ETH Zürich(苏黎世联邦理工学院);California Institute of Technology(加州理工学院);University of Alberta(阿尔伯塔大学);Alberta Machine Intelligence Institute (Amii)(阿尔伯塔人工智能研究所);Cedars-Sinai Medical Center(Cedars-Sinai 医疗中心);Archimedes AI, Athena Research Center(Archimedes AI,Athena 研究中心)
Sample-Efficient Reinforcement Learning Controller for Deep Brain Stimulation in Parkinson's Disease
Department of Electrical and Computer Engineering, Clemson University(电子与计算机工程系,克莱姆森大学);Department of Biomedical Engineering, Michigan Technological University(生物医学工程系,密歇根技术大学)