4 ms·
>> I am making an inference here, but I imagine his complaint is with the lack of ability to derive A and B in your "If A, then B" example. Full disclosure: I
by YeGoblynQueenne 9y ago
>> I am making an inference here, but I imagine his complaint is with the lack of ability to derive A and B in your "If A, then B" example.
Full disclosure: I work with Meta-Interpretive Learning, a class of algorithms that can do exactly that, for my PhD. See [1] for an overview and [2] for a more detailed explanation (I am not an author in any of those papers). MIL can go a lot further than deriving implications in propositional logic- it learns first-order logic theories, which should cover professor Pearl's expectations quite adequately.
For instance, in a MIL setting it's perfectly possible to learn a theory for e.g. cause_of(A,B), from examples of what A causes what B.
As to learning counterfactuals, my thesis advisor has published papers on a Robot Scientist, a machine that carries out scientific experiments from start to end: it forms hypotheses and carries out experiments to test them, then refines its theories, etc [3].
That last paper was published in Nature- which makes it even harder to see how Pearl can claim that nobody knows how to do this sort of thing in machine learning. I understand if people just want to ride on the coattails of deep learning and only talk about that in the press, but it's a bit annoying all the same, to see well-established results ignored by someone so learned.
___________
[1] Meta-Interpretive Learning: achievements and challenges
https://www.doc.ic.ac.uk/~shm/Papers/rulemlabs.pdf https://www.doc.ic.ac.uk/~shm/Papers/rulemlabs.pdf
[2] Meta-interpretive learning of higher-order dyadic datalog: predicate invention revisited
https://www.doc.ic.ac.uk/~shm/Papers/metagolDMLJ.pdf https://www.doc.ic.ac.uk/~shm/Papers/metagolDMLJ.pdf
[3] Functional genomic hypothesis generation and experimentation by a robot scientist
https://www.nature.com/articles/nature02236 https://www.nature.com/articles/nature02236
- jamez1 9y agoThat's interesting, thanks for the links. I have mucked around with program induction using expression trees but had not come across MIL before. It looks like MIL would generalize quite well to bayesian networks, a perfect bridge between his past research and what he's depicting as the future of ML research.