ML Seminar: Alexander Marx (International Max Planck Research School for Informatics)

Everyone is welcome to attend de ML seminar of Alexander Marx with the title 'Testing Conditional Independence on Discrete Data using Stochastic Complexity'.
  • What English Machine Learning Seminars
  • When 10-10-2019 from 11:00 to 12:00 (Europe/Amsterdam / UTC200)
  • Where Room L016 at CWI, Science Park 123 in Amsterdam
  • Contact Name Wouter Koolen-Wijkstra
  • Add event to calendar iCal

Everyone is welcome to attend de ML seminar of Alexander Marx with the title 'Testing Conditional Independence on Discrete Data using Stochastic Complexity'.

Abstract: Testing for conditional independence is a core aspect of constraint-based causal discovery. Although commonly used tests are
perfect in theory, they often fail to reject independence in practice, especially when conditioning on multiple variables.
We focus on discrete data and propose a new test based on the notion of algorithmic independence that we instantiate using stochastic
complexity. Amongst others, we show that our proposed test, SCI, is an asymptotically unbiased as well as L2 consistent estimator for
conditional mutual information (CMI). Further, we show that SCI can be reformulated to find a sensible threshold for CMI that works well on limited samples. Empirical evaluation shows that SCI has a lower type II error than commonly used tests. As a result, we obtain a higher recall when we use SCI in causal discovery algorithms, without compromising the precision.