Following the advice of doing short blogposts, let’s give it ago.

Currently I am:

Organising IAI CDT Spring Research Conference

Organising this has been both fun (we get to choose some of the stuff that happens) and frustrating (when a speaker we ask isn’t available on the dates; worrying the other CDT students think it’s all a waste of time). The main purpose of it is for students of the CDT to present our research to each other but we’ve got a couple of interesting speakers coming and hopefully we’ve have some fun too.

Conferences

Looking forward to finally getting to a conference in-person (Covid and caring responsibilities make this not so simple) but not really sure what to expect at a scientific conference and how to make it useful.

Improving my diffusion model

Bit overwhelmed by prioritising and getting through this (not exhaustive) list:

  • Quantile Mapping
  • Working out why my RAPSD plots are weird
  • Evaluating the probalistic part of ML model
  • Climate change signal response
  • Communicating variability in QQ plots and bias (so variability in CPM and sample quantiles over all locations, CPM and sample means at each location)
  • Puttingn together a paper
  • Figuring what the hell will my thesis contain
  • Better way to display samples
  • Picking and fitting some decent baselines (GAN, deterministic UNet, GLMs that perhaps can’t do spatially coherent single samples)