BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20251122T033319EST-6773TgNA2Z@132.216.98.100 DTSTAMP:20251122T083319Z DESCRIPTION:Title: Optimal One-pass Nonparametric Estimation Under Memory C onstraint.\n\n\n Abstract:\n\n\nFor nonparametric regression in the streami ng setting\, where data constantly flow in and require real-time analysis\ , a main challenge is that data are cleared from the computer system once processed due to limited computer memory and storage. We tackle the challe nge by proposing a novel one-pass estimator based on penalized orthogonal basis expansions and developing a general framework to study the interplay between statistical efficiency and memory consumption of estimators. We s how that\, the proposed estimator is statistically optimal under memory co nstraint\, and has asymptotically minimal memory footprints among all one- pass estimators of the same estimation quality. Numerical studies demonstr ate that the proposed one-pass estimator is nearly as efficient as its non -streaming counterpart that has access to all historical data.\n\n\n Speake r\n\n\nZhenhua Lin is a Presidential Young Professor in the Department of Statistics and Data Science at National University of Singapore. He is als o an Affiliated Faculty in the Institute of Data Science. His research int erests include Functional data analysis non-Euclidean data analysis\, high -dimensional data analysis\, statistics under non-statistical constraints. He is the Associate Editor of Bernoulli Journal.\n\nhttps://blog.nus.edu. sg/zhenhua/\n\nɬÀï·¬ Statistics Seminar schedule: https://mcgillstat.gith ub.io/\n\nZoom meeting\n\nhttps://mcgill.zoom.us/j/83436686293?pwd=b0RmWml XRXE3OWR6NlNIcWF5d0dJQT09\n\nMeeting ID: 834 3668 6293\n\nPasscode: 12345 \n\n\n \n  \n  \n \n \n\n DTSTART:20221118T203000Z DTEND:20221118T213000Z SUMMARY:Zhenhua Lin (National University of Singapore) URL:/mathstat/channels/event/zhenhua-lin-national-univ ersity-singapore-343540 END:VEVENT END:VCALENDAR