{"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/data-driven-perception-of-neuron-point","title":"Data-driven Perception of Neuron Point Process with Unknown Unknowns","arxiv_id":"1811.00688","date":"2018-11-02","proceeding":null,"authors":["Ruochen Yang","Gaurav Gupta","Paul Bogdan"],"abstract":"Identification of patterns from discrete data time-series for statistical\ninference, threat detection, social opinion dynamics, brain activity prediction\nhas received recent momentum. In addition to the huge data size, the associated\nchallenges are, for example, (i) missing data to construct a closed\ntime-varying complex network, and (ii) contribution of unknown sources which\nare not probed. Towards this end, the current work focuses on statistical\nneuron system model with multi-covariates and unknown inputs. Previous research\nof neuron activity analysis is mainly limited with effects from the spiking\nhistory of target neuron and the interaction with other neurons in the system\nwhile ignoring the influence of unknown stimuli. We propose to use unknown\nunknowns, which describes the effect of unknown stimuli, undetected neuron\nactivities and all other hidden sources of error. The maximum likelihood\nestimation with the fixed-point iteration method is implemented. The\nfixed-point iterations converge fast, and the proposed methods can be\nefficiently parallelized and offer computational advantage especially when the\ninput spiking trains are over long time-horizon. The developed framework\nprovides an intuition into the meaning of having extra degrees-of-freedom in\nthe data to support the need for unknowns. The proposed algorithm is applied to\nsimulated spike trains and on real-world experimental data of mouse\nsomatosensory, mouse retina and cat retina. The model shows a successful\nincreasing of system likelihood with respect to the conditional intensity\nfunction, and it also reveals the convergence with iterations. Results suggest\nthat the neural connection model with unknown unknowns can efficiently estimate\nthe statistical properties of the process by increasing the network likelihood.","url_abs":"http://arxiv.org/abs/1811.00688v2","url_pdf":"http://arxiv.org/pdf/1811.00688v2.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":"data-driven-perception-of-neuron-point","repo_url":"https://github.com/gaurav71531/spikeNetwork","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"activity-prediction","task_name":"Activity Prediction"},{"task_slug":"time-series","task_name":"Time Series Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}