skip to content

MSG Design of Experiments Seminar Series: The war against bias: experimental design for big data

Presented by: 
Henry Wynn London School of Economics
Wednesday 20th June 2018 - 14:05 to 14:55
INI Seminar Room 1
The talk first reviews  work (by others) on optimal experimental design for “big data”. This ranges from methods arising from the social and medical  sciences, particularly in causal modelling, to recent work which tries to extract an optimum design from a loosely structured data set of covariates and also the literature on optimal design to guard against bias. The authors draw on some of  this work but take a more game-theoretic approach. The idea is that the causal modelling operation, run by an notional “Alice”, needs a shield protecting against bias built by a notional “Bob”. The two operation can act  harmoniously  when the joint operation is a over product space but, even when not,  a Nash equilibrium may be achievable, which balances the two objectives.

Joint work with Elena Pesce and Eva Riccomagno.
The video for this talk should appear here if JavaScript is enabled.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.
University of Cambridge Research Councils UK
    Clay Mathematics Institute London Mathematical Society NM Rothschild and Sons