Machine learning assumes data to be independently and identically sampled from a distribution, though in the real world this assumption rarely holds. Data stream mining, a way of tackling data streams that change over time, is introduced in "Everything Changes, but Your ML Models Stay the Same?", a talk by Bernhard Pfahringer of the University of Waikato. The talk is part of OFAI's 2022 Lecture Series.
Members of the public are cordially invited to attend the talk via Zoom on Wednesday, 16 November at 19:30 CET (UTC+1):
Meeting ID: 842 8244 2460
(Note that this talk has been rescheduled from its originally announced slot of Wednesday, 9 November at 18:30 CET.)
Talk abstract: Most Machine Learning assumes i.i.d. data, data that is independently and identically sampled from a distribution. Thus training on a sample and then applying to new samples is a sound procedure. Unfortunately, in the real world, data is almost never i.i.d. This presentation will introduce data stream mining as a way of tackling data streams that change over time. I will in particular highlight opportunities that this scenario offers, that are not present in the static i.i.d. train-then-test setup.
Speaker biography: Bernhard Pfahringer is a professor with the Computer Science Department at the University of Waikato, and also a co-director for its new AI Institute.