BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Nina Miolane (UC Santa Barbara) DTSTART:20220601T180000Z DTEND:20220601T184500Z DTSTAMP:20260423T050805Z UID:SageDays112358/8 DESCRIPTION:Title: Geomstats – a Python package for differential geometry in statist ics and machine learning\nby Nina Miolane (UC Santa Barbara) as part o f Global Virtual SageDays 112.358\n\n\nAbstract\nWe introduce Geomstats\, an open-source Python package for computations and statistics on nonlinear manifolds that appear in machine learning applications\, such as: hyperbo lic spaces\, spaces of symmetric positive definite matrices\, Lie groups o f transformations\, and many more. We provide object-oriented and extensiv ely unit-tested implementations. Manifolds come equipped with families of Riemannian metrics with associated exponential and logarithmic maps\, geod esics\, and parallel transport. Statistics and learning algorithms provide methods for estimation\, clustering\, and dimension reduction on manifold s. All associated operations provide support for different execution backe nds — namely NumPy\, Autograd\, PyTorch\, and TensorFlow. This talk pres ents the package\, compares it with related libraries\, and provides relev ant examples. We show that Geomstats provides reliable building blocks to both foster research in differential geometry and statistics and democrati ze the use of (Riemannian) geometry in statistics and machine learning. Th e source code is freely available under the MIT license at https://github. com/geomstats/geomstats.\n LOCATION:https://researchseminars.org/talk/SageDays112358/8/ END:VEVENT END:VCALENDAR