An anatomy-based V1 model: Extraction of Low-level Features, Reduction of distortion and a V1-inspired SOM

We present a model of the primary visual cortex V1, guided by anatomical experiments. Unlike most machine learning systems our goal is not to maximize accuracy but to realize a system more aligned to biological systems. Our model consists of the V1 layers 4, 2/3, and 5, with inter-layer connections between them in accordance with the anatomy. We further include the orientation selectivity of the V1 neurons and lateral influences in each layer. Our V1 model, when applied to the BSDS500 ground truth images (indicating LGN contour detection before V1), can extract low-level features from the images and perform a significant amount of distortion reduction. As a follow-up to our V1 model, we propose a V1-inspired self-organizing map algorithm (V1-SOM), where the weight update of each neuron gets influenced by its neighbors. V1-SOM can tolerate noisy inputs as well as noise in the weight updates better than SOM and shows a similar level of performance when trained with high dimensional data such as the MNIST dataset. Finally, when we applied V1 processing to the MNIST dataset to extract low-level features and trained V1-SOM with the modified MNIST dataset, the quantization error was significantly reduced. Our results support the hypothesis that the ventral stream performs gradual untangling of input spaces.

[1]  Eero P. Simoncelli,et al.  Toward Next-Generation Artificial Intelligence: Catalyzing the NeuroAI Revolution , 2022, ArXiv.

[2]  Megan B. Kratz,et al.  Synaptic connectivity to L2/3 of primary visual cortex measured by two-photon optogenetic stimulation , 2022, eLife.

[3]  C. Clopath,et al.  Excitatory-inhibitory balance modulates the formation and dynamics of neuronal assemblies in cortical networks , 2021, bioRxiv.

[4]  A. Angelucci,et al.  Anatomy and Physiology of Macaque Visual Cortical Areas V1, V2, and V5/MT: Bases for Biologically Realistic Models. , 2020, Cerebral cortex.

[5]  Elias B. Issa,et al.  Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs , 2019, NeurIPS.

[6]  Saumil S. Patel,et al.  Layer 4 of mouse neocortex differs in cell types and circuit organization between sensory areas , 2019, Nature Communications.

[7]  Akinari Onishi,et al.  Landmark Map: An Extension of the Self-Organizing Map for a User-Intended Nonlinear Projection , 2019, Neurocomputing.

[8]  James J. DiCarlo,et al.  Reversible Inactivation of Different Millimeter-Scale Regions of Primate IT Results in Different Patterns of Core Object Recognition Deficits , 2018, Neuron.

[9]  Pozi Milow,et al.  Random forest and Self Organizing Maps application for analysis of pediatric fracture healing time of the lower limb , 2018, Neurocomputing.

[10]  Ján Antolík,et al.  Rapid Long-Range Disynaptic Inhibition Explains the Formation of Cortical Orientation Maps , 2017, Front. Neural Circuits.

[11]  Xiaolin Wang,et al.  SOMH: A self-organizing map based topology preserving hashing method , 2016, Neurocomputing.

[12]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  Alexander S. Ecker,et al.  Principles of connectivity among morphologically defined cell types in adult neocortex , 2015, Science.

[14]  Everton J. Agnes,et al.  Diverse synaptic plasticity mechanisms orchestrated to form and retrieve memories in spiking neural networks , 2015, Nature Communications.

[15]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[16]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[17]  George Azzopardi,et al.  A Push-Pull CORF Model of a Simple Cell with Antiphase Inhibition Improves SNR and Contour Detection , 2014, PloS one.

[18]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[19]  George Azzopardi,et al.  A CORF computational model of a simple cell that relies on LGN input outperforms the Gabor function model , 2012, Biological Cybernetics.

[20]  James J. DiCarlo,et al.  How Does the Brain Solve Visual Object Recognition? , 2012, Neuron.

[21]  Charless C. Fowlkes,et al.  Contour Detection and Hierarchical Image Segmentation , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  Sooyoung Chung,et al.  Highly ordered arrangement of single neurons in orientation pinwheels , 2006, Nature.

[23]  Stephen D. Van Hooser,et al.  Orientation Selectivity without Orientation Maps in Visual Cortex of a Highly Visual Mammal , 2005, The Journal of Neuroscience.

[24]  Bhabatosh Chanda,et al.  Design of vector quantizer for image compression using self-organizing feature map and surface fitting , 2004, IEEE Transactions on Image Processing.

[25]  J. Lübke,et al.  Morphometric analysis of the columnar innervation domain of neurons connecting layer 4 and layer 2/3 of juvenile rat barrel cortex. , 2003, Cerebral cortex.

[26]  T. Bonhoeffer,et al.  Mapping Retinotopic Structure in Mouse Visual Cortex with Optical Imaging , 2002, The Journal of Neuroscience.

[27]  Arnaud Delorme,et al.  Feed-forward contour integration in primary visual cortex based on asynchronous spike propagation , 2001, Neurocomputing.

[28]  R. Lund,et al.  Receptive field properties of single neurons in rat primary visual cortex. , 1999, Journal of neurophysiology.

[29]  Klaus Obermayer,et al.  Effects of lateral competition in the primary visual cortex on the development of topographic projections and ocular dominance maps , 1999, Neurocomputing.

[30]  Risto Miikkulainen,et al.  Laterally Interconnected Self-Organizing Maps in Hand-Written Digit Recognition , 1995, NIPS.

[31]  James C. Bezdek,et al.  A note on self-organizing semantic maps , 1995, IEEE Trans. Neural Networks.

[32]  Olvi L. Mangasarian,et al.  Nuclear feature extraction for breast tumor diagnosis , 1993, Electronic Imaging.

[33]  Amiram Grinvald,et al.  Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns , 1991, Nature.

[34]  G. Blasdel,et al.  Voltage-sensitive dyes reveal a modular organization in monkey striate cortex , 1986, Nature.

[35]  Kunihiko Fukushima,et al.  Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position , 1980, Biological Cybernetics.

[36]  N. Berman,et al.  The rabbit and the cat: A comparison of some features of response properties of single cells in the primary visual cortex , 1979, The Journal of comparative neurology.

[37]  C. Blakemore,et al.  Functional organization in the visual cortex of the golden hamster , 1976, The Journal of comparative neurology.

[38]  D H HUBEL,et al.  RECEPTIVE FIELDS AND FUNCTIONAL ARCHITECTURE IN TWO NONSTRIATE VISUAL AREAS (18 AND 19) OF THE CAT. , 1965, Journal of neurophysiology.

[39]  Pieter R. Roelfsema,et al.  Attention-Gated Brain Propagation: How the brain can implement reward-based error backpropagation , 2020, NeurIPS.

[40]  Jenna G. Kelly,et al.  Major Feedforward Thalamic Input Into Layer 4C of Primary Visual Cortex in Primate , 2019, Cerebral cortex.

[41]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[42]  M. Imbert,et al.  The primary visual cortex in the mouse: Receptive field properties and functional organization , 2004, Experimental Brain Research.

[43]  Keiji Tanaka.  Columns for complex visual object features in the inferotemporal cortex: clustering of cells with similar but slightly different stimulus selectivities. , 2003, Cerebral cortex.

[44]  Wulfram Gerstner,et al.  Spiking Neuron Models , 2002 .

[45]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[46]  D. Hubel,et al.  Receptive fields, binocular interaction and functional architecture in the cat's visual cortex , 1962, The Journal of physiology.