{"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/putting-a-bug-in-ml-the-moth-olfactory","title":"Putting a bug in ML: The moth olfactory network learns to read MNIST","arxiv_id":"1802.05405","date":"2018-02-15","proceeding":null,"authors":["Charles B. Delahunt","J. Nathan Kutz"],"abstract":"We seek to (i) characterize the learning architectures exploited in\nbiological neural networks for training on very few samples, and (ii) port\nthese algorithmic structures to a machine learning context. The Moth Olfactory\nNetwork is among the simplest biological neural systems that can learn, and its\narchitecture includes key structural elements and mechanisms widespread in\nbiological neural nets, such as cascaded networks, competitive inhibition, high\nintrinsic noise, sparsity, reward mechanisms, and Hebbian plasticity. These\nstructural biological elements, in combination, enable rapid learning.\n  MothNet is a computational model of the Moth Olfactory Network, closely\naligned with the moth's known biophysics and with in vivo electrode data\ncollected from moths learning new odors. We assign this model the task of\nlearning to read the MNIST digits. We show that MothNet successfully learns to\nread given very few training samples (1 to 10 samples per class). In this\nfew-samples regime, it outperforms standard machine learning methods such as\nnearest-neighbors, support-vector machines, and neural networks (NNs), and\nmatches specialized one-shot transfer-learning methods but without the need for\npre-training. The MothNet architecture illustrates how algorithmic structures\nderived from biological brains can be used to build alternative NNs that may\navoid some of the learning rate limitations of current engineered NNs.","url_abs":"http://arxiv.org/abs/1802.05405v3","url_pdf":"http://arxiv.org/pdf/1802.05405v3.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":"putting-a-bug-in-ml-the-moth-olfactory","repo_url":"https://github.com/charlesDelahunt/PuttingABugInML","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"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}