BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Daniele Angella (Università degli Studi di Firenze) DTSTART:20240206T143000Z DTEND:20240206T153000Z DTSTAMP:20260507T233135Z UID:DifferentialAndAlgebraicGeometry/22 DESCRIPTION:Title: Constructing and Machine Learning Calabi-Yau Fiv e-Folds\nby Daniele Angella (Università degli Studi di Firenze) as pa rt of Geometry in Como\n\n\nAbstract\nThe significance of Calabi-Yau manif olds transcends both Complex Geometry and String Theory.\nOne possible app roach to constructing Calabi-Yau manifolds involves intersecting hypersurf aces within the product of projective spaces\, defined by polynomials of a specific degree.\nWe show a method to construct all possible complete int ersections Calabi-Yau five-folds within a product of four or less complex projective spaces\, with up to four constraints. This results in a compre hensive set of 27\,068 distinct spaces.\nFor approximately half of these c onstructions\, excluding the product spaces\, we can compute the cohomolog ical data\, yielding 2\,375 distinct Hodge diamonds.\nWe present distribut ions of the invariants and engage in a comparative analysis with their low er-dimensional counterparts.\nSupervised machine learning techniques are a pplied to the cohomological data. The Hodge number $h^{1\,1}$ can be learn t with high efficiency\; however\, accuracy diminishes for other Hodge num bers due to the extensive ranges of potential values.\n\nThe talk is a joi nt collaboration with Rashid Alawadhi\, Andrea Leonardo\, and Tancredi Sch ettini Gherardini.\n LOCATION:https://researchseminars.org/talk/DifferentialAndAlgebraicGeometr y/22/ END:VEVENT END:VCALENDAR