{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/correspondence-of-deep-neural-networks-and","title":"Correspondence of Deep Neural Networks and the Brain for Visual Textures","arxiv_id":"1806.02888","date":"2018-06-07","proceeding":null,"authors":["Md Nasir Uddin Laskar","Luis G. Sanchez Giraldo","Odelia Schwartz"],"abstract":"Deep convolutional neural networks (CNNs) trained on objects and scenes have\nshown intriguing ability to predict some response properties of visual cortical\nneurons. However, the factors and computations that give rise to such ability,\nand the role of intermediate processing stages in explaining changes that\ndevelop across areas of the cortical hierarchy, are poorly understood. We\nfocused on the sensitivity to textures as a paradigmatic example, since recent\nneurophysiology experiments provide rich data pointing to texture sensitivity\nin secondary but not primary visual cortex. We developed a quantitative\napproach for selecting a subset of the neural unit population from the CNN that\nbest describes the brain neural recordings. We found that the first two layers\nof the CNN showed qualitative and quantitative correspondence to the cortical\ndata across a number of metrics. This compatibility was reduced for the\narchitecture alone rather than the learned weights, for some other related\nhierarchical models, and only mildly in the absence of a nonlinear computation\nakin to local divisive normalization. Our results show that the CNN class of\nmodel is effective for capturing changes that develop across early areas of\ncortex, and has the potential to facilitate understanding of the computations\nthat give rise to hierarchical processing in the brain.","url_abs":"http://arxiv.org/abs/1806.02888v1","url_pdf":"http://arxiv.org/pdf/1806.02888v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"correspondence-of-deep-neural-networks-and","repo_url":"https://github.com/nasirml/DeepNetAndBrain","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"sensitivity","task_name":"Sensitivity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.02888","atlas_url":"https://app.syntology.ai/?focus=1806.02888","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}