BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:CD Technical Meeting (ML15): Learning compact space representation
  of plasma states from the FairMAST dataset
DTSTART:20260730T100000Z
DTEND:20260730T110000Z
DTSTAMP:20260819T234000Z
UID:indico-event-803@indico.ukaea.uk
DESCRIPTION:Speakers: Andrea Loreti\n\nMachine Learning\, Uncertainty Quan
 tification and Data Science \n \nThe FairMAST dataset provides a unique 
 machine-learning-ready collection of more than 11\,000 MAST discharges\, i
 ncluding magnetic\, equilibrium\, current\, voltage and auxiliary diagnost
 ic measurements. However\, the data present significant challenges due to 
 missing signals\, heterogeneous diagnostic modalities and widely varying s
 ampling frequencies. \nThis work presents a Variational Autoencoder (VAE)
  framework designed to learn compact latent-space representations of multi
 modal plasma diagnostics while remaining robust to sparse and incomplete m
 easurements. \nUsing magnetic diagnostics from FairMAST\, the model is ap
 plied to both equilibrium reconstruction and short-term forecasting tasks\
 , demonstrating accurate recovery of EFIT-derived quantities\, including p
 oloidal flux maps and plasma boundary information.\n \n\nhttps://indico.u
 kaea.uk/event/803/
URL:https://indico.ukaea.uk/event/803/
END:VEVENT
END:VCALENDAR