IARPA Geopolitical Forecasting Challenge 2

IARPA is the US Intelligence Advanced Research Projects Activity (The Intelligence equivalent of DARPA) and its aim is to:

stimulate breakthroughs in the science of forecasting to create greater strategic advantages for maintaining global security, predicting economic trends and directing the need for humanitarian efforts.

The 1st Geopolitical Forecasting Challenge was partly covered in the book Superforecasting. This 2nd Challenge was more challenging in three main ways:

  • Larger volume of events to be forecast (c 300 in total), so requires use of algorithms rather than relying on ad-hoc analysis
  • Had to Forecast against all the problems set, not just picking and choosing.
  • Focus on use of IT based automated systems to scale prediction capability.

The Challenge ran between May and November 2019 and required participants to forecast a large number of different events – political, financial, macroeconomic, technological, diseases, disasters, conflicts etc (for example – “Will the UK leave the EU by November the 1st”)

DataSwarm Involvement

DataSwarm entered the competition with a Beta version of what became the first version of our Zeitgeist General Prediction System. We won one of the minor prizes in the tournament.

An interesting outcome was the top 5 places were all won by expert “Geopolitics 1” forecasters using some automation – this has led to a lot of research in hybrid human/machine forecasting by IARPA and many others.

Lessons Learned

We have taken a lot of this IARPA learning, and together with our work in building the DataSwarm Markets asset prediction system, ew have used it to build the upgraded second version of the Zeitgeist General Prediction System.