Cause-effect pairs challenge

Post date: May 7, 2013 5:40:44 PM

Isabelle Guyon

ChaLearn

http://www.causality.inf.ethz.ch/cause-effect.php

The problem of attributing causes to effects is pervasive in science, medicine, economy and almost every aspects of our everyday life involving human reasoning and decision making. What affects your health? the economy? climate changes? The gold standard to establish causal relationships is to perform randomized controlled experiments. However, experiments are costly while non-experimental "observational" data collected routinely around the world are readily available. Unraveling potential cause-effect relationships from such observational data could save a lot of time and effort.

Consider for instance a target variable B, like occurrence of "lung cancer" in patients. The goal would be to find whether a factor A, like "smoking", might cause B. The objective of the challenge is to rank pairs of variables {A, B} to prioritize experimental verifications of the conjecture that A causes B.

As is known, "correlation does not mean causation". More generally, observing a statistical dependency between A and B does not imply that A causes B or that B causes A; A and B could be consequences of a common cause. But, is it possible to determine from the joint observation of samples of two variables A and B that A should be a cause of B? There are new algorithms that have appeared in the literature in the past few years that tackle this problem. This challenge is an opportunity to evaluate them and propose new techniques to improve on them.

We provide hundreds of pairs of real variables with known causal relationships from domains as diverse as chemistry, climatology, ecology, economy, engineering, epidemiology, genomics, medicine, physics. and sociology. Those are intermixed with controls (pairs of independent variables and pairs of variables that are dependent but not causally related) and semi-artificial cause-effect pairs (real variables mixed in various ways to produce a given outcome).

This challenge is limited to pairs of variables deprived of their context. Thus constraint-based methods relying on conditional independence tests and/or graphical models are not applicable. The goal is to push the state-of-the art in complementary methods, which can eventually disambiguate Markov equivalence classes.

This project organized by ChaLearn is sponsored by Pascal2 EU network of excellence and is part of the official selection of the IJCNN 2013 competitions. We are grateful to ETH Zurich and Kaggle for hosting the websites and to the many volunteers who contributed to the organization. The principal co-organizers are Ben Hamner (Kaggle), Alexander Statnikov (NYU), Mikael Henaff (NYU), Vincent Lemaire (Orange), and Bernhard Shoelkopf (MPI).