{"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/a-neurobiological-evaluation-metric-for","title":"A Neurobiological Evaluation Metric for Neural Network Model Search","arxiv_id":"1805.10726","date":"2018-05-28","proceeding":"CVPR 2019 6","authors":["Nathaniel Blanchard","Jeffery Kinnison","Brandon RichardWebster","Pouya Bashivan","Walter J. Scheirer"],"abstract":"Neuroscience theory posits that the brain's visual system coarsely identifies\nbroad object categories via neural activation patterns, with similar objects\nproducing similar neural responses. Artificial neural networks also have\ninternal activation behavior in response to stimuli. We hypothesize that\nnetworks exhibiting brain-like activation behavior will demonstrate brain-like\ncharacteristics, e.g., stronger generalization capabilities. In this paper we\nintroduce a human-model similarity (HMS) metric, which quantifies the\nsimilarity of human fMRI and network activation behavior. To calculate HMS,\nrepresentational dissimilarity matrices (RDMs) are created as abstractions of\nactivation behavior, measured by the correlations of activations to stimulus\npairs. HMS is then the correlation between the fMRI RDM and the neural network\nRDM across all stimulus pairs. We test the metric on unsupervised predictive\ncoding networks, which specifically model visual perception, and assess the\nmetric for statistical significance over a large range of hyperparameters. Our\nexperiments show that networks with increased human-model similarity are\ncorrelated with better performance on two computer vision tasks: next frame\nprediction and object matching accuracy. Further, HMS identifies networks with\nhigh performance on both tasks. An unexpected secondary finding is that the\nmetric can be employed during training as an early-stopping mechanism.","url_abs":"http://arxiv.org/abs/1805.10726v4","url_pdf":"http://arxiv.org/pdf/1805.10726v4.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":"a-neurobiological-evaluation-metric-for","repo_url":"https://github.com/CVRL/human-model-similarity","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}