(2009-09-21) Swartz Pearl Causality Social Science

Aaron Swartz describes Judea Pearl's attempts to use AI techniques to evaluate causality, not just correlation, with Social Science Statistics. One of the reasons I find these areas to be fertile grounds to try out new ideas is that, unlike AI, tangible rewards can be reaped from solving relative small problems. Problems involving barely 4 to 5 variables, which we in AI regard as toy-problems, carry tremendous payoffs in public health and social science.