Joe Roussos

Joe Roussos

I am a PhD student in philosophy at the London School of Economics, working with Roman Frigg and Richard Bradley.

I work on two related fields: (1) decision-making under conditions of severe uncertainty, and (2) scientific modelling, especially as a source of evidence for decisions. My recent work has focused on climate science, and in particular on the prediction of hurricanes striking North America.

More broadly, I am interested in the social epistemology of science: how it should be organised, and how it should interact with the rest of society. Much of my work can be characterised as formal epistemology (i.e., I use mathematical tools), though much of my interest is in non-ideal, messy situations in which our favoured formal tools fare poorly. 

A copy of my CV is available here.

I help organise the LSE’s Conjectures and Refutations seminar series.

Before beginning my PhD I worked as a management consultant for Bain & Co., and then in the strategy team for London-based property developers Lodha UK. I have a Master’s in theoretical physics from the University of Cambridge, and my undergraduate degree was in Physics and Philosophy at the University of the Witwatersrand.

I also work as a freelance consultant, on strategy, management and decision-making.


Recent talks

  • “Making confident decisions with ensembles” – EPSA 2019 Geneva [slides]
  • “Formal epistemology as modelling” – PROGIC 2019 Frankfurt 
  • “Formal Epistemology as Modelling” – FEW 2019 Turin [poster]
  • “Making confident decisions with models” – CHESS Durham [slides]
  • “Scoring rules and expert disagreement” – Masterclass on social epistemology Glasglow [handout]


Work in progress papers

  • Making confident decisions with model ensembles (R&R)
  • Formal epistemology as normative modelling (draft)
  • Expert disagreement: aggregation and scoring rules for real agents (draft)