Theoretical Impediments to Machine Learning A position paper

@inproceedings{Pearl2016TheoreticalIT,
  title={Theoretical Impediments to Machine Learning A position paper},
  author={Judea Pearl},
  year={2016}
}
Current machine learning systems operate, almost exclusively, in a purely statistical mode, which puts severe theoretical limits on their performance. We consider the feasibility of leveraging counterfactual reasoning in machine learning tasks, and to identify areas where such reasoning could lead to major breakthroughs in machine learning applications… CONTINUE READING