BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Stochastic Parameterizations: Better Modelling of Temporal Correla
 tions using ML - Raghul Parthipan\, University of Cambridge
DTSTART:20220405T100000Z
DTEND:20220405T113000Z
UID:TALK172130@talks.cam.ac.uk
CONTACT:Herbie Bradley
DESCRIPTION:The modelling of small-scale processes is a major source of er
 ror in climate models\, hindering the accuracy of low-cost models which mu
 st approximate such processes through parameterization. Using stochasticit
 y and machine learning have led to better models but there is a lack of wo
 rk on combining the benefits from both. We show that by using a physically
 -informed recurrent neural network within a probabilistic framework\, our 
 resulting model for the Lorenz 96 atmospheric simulation is competitive an
 d often superior to both a bespoke baseline and an existing probabilistic 
 machine-learning (GAN) one. This is due to a superior ability to model tem
 poral correlations compared to standard first-order autoregressive schemes
 . The model also generalises to unseen regimes. We evaluate across a numbe
 r of metrics from the literature\, but also discuss how the probabilistic 
 metric of likelihood may be a unifying choice for future probabilistic cli
 mate models.
LOCATION:https://zoom.us/j/6708259482?pwd=Qk03U3hxZWNJZUZpT2pVZnFtU2RRUT09
END:VEVENT
END:VCALENDAR