{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Goldin MA"],"funding":["EC | Horizon Europe | Excellent Science | HORIZON EUROPE Marie Sklodowska-Curie Actions","Agence Nationale de la Recherche","Fondation de France"],"pagination":["e2301150120"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC10450671"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["120(34)"],"pubmed_abstract":["Predicting the responses of sensory neurons is a long-standing neuroscience goal. However, while there has been much progress in modeling neural responses to simple and/or artificial stimuli, predicting responses to natural stimuli remains an ongoing challenge. On the one hand, deep neural networks perform very well on certain datasets but can fail when data are limited. On the other hand, Gaussian processes (GPs) perform well on limited data but are poor at predicting responses to high-dimensional stimuli, such as natural images. Here, we show how structured priors, e.g., for local and smooth receptive fields, can be used to scale up GPs to model neural responses to high-dimensional stimuli. With this addition, GPs largely outperform a deep neural network trained to predict retinal respon"],"journal":["Proceedings of the National Academy of Sciences of the United States of America"],"pubmed_title":["Scalable Gaussian process inference of neural responses to natural images."],"pmcid":["PMC10450671"],"funding_grant_id":["OPC - 17-CE37-0013-01","00120244","861423"],"pubmed_authors":["Goldin MA","Chalk M","Virgili S"],"additional_accession":[]},"is_claimable":false,"name":"Scalable Gaussian process inference of neural responses to natural images.","description":"Predicting the responses of sensory neurons is a long-standing neuroscience goal. However, while there has been much progress in modeling neural responses to simple and/or artificial stimuli, predicting responses to natural stimuli remains an ongoing challenge. On the one hand, deep neural networks perform very well on certain datasets but can fail when data are limited. On the other hand, Gaussian processes (GPs) perform well on limited data but are poor at predicting responses to high-dimensional stimuli, such as natural images. Here, we show how structured priors, e.g., for local and smooth receptive fields, can be used to scale up GPs to model neural responses to high-dimensional stimuli. With this addition, GPs largely outperform a deep neural network trained to predict retinal respon","dates":{"release":"2023-01-01T00:00:00Z","publication":"2023 Aug","modification":"2025-04-18T20:56:49.868Z","creation":"2025-04-07T08:59:35.563Z"},"accession":"S-EPMC10450671","cross_references":{"pubmed":["37579153"],"doi":["10.1073/pnas.2301150120"]}}