Learning Explanatory Rules from Noisy Data · HackerPapers discussions

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Sun, 12 Sep 2021 19:29:03 GMT

You're not too far off. The big advantage of ILP with respect to statistical techniques is the really extraordinary sample efficiency. Just a couple of examples are enough to learn many programs [1]. Moreover, ILP is pretty much the only machine learning approach that can make use of background knowledge [2] and in fact is characterised by this ability, and it is the ability to use background knowledge that is responsible for the excellent sample complexity, because if you don't have any background knowledge you are forced to learn everything from scratch - as in the "end-to-end" approach…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Wed, 11 Nov 2020 00:35:10 GMT

Aren't you shifting the goalposts a bit? First you ask me how I'd solve this problem with ILP: “Given three strings n1, n2, and n3, return true if length(n1) < length(n2) < length(n3), and false otherwise” (a) I say this is a trivial problem to solve with ILP and I show you a, well, trivial solution and you complain that it's - trivial. Then I show you a more elaborate version that learns sub-programs on the way to the full solution and you say that a) it doesn't solve a different problem, with ≤ instead of < and b) that it's not trivial anymore. And now you're saying you want a solution that…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by sanxiyn

Tue, 04 Jun 2019 04:37:33 GMT

Inductive Logic Programming. DeepMind work in question is https://arxiv.org/abs/1711.04574.

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Mon, 03 Jun 2019 06:40:40 GMT

Hi. ILP is Inductive Logic Programming, a form of logic-based, symbolic machine learning. ILP "models" are first-order logic theories that are not differentiable. To put it plainly, most ILP algorithms learn Prolog programs from examples and background knowledge which are also Prolog programs. Some learn logic programs in other logic programming languages, like Answer Set Programming or constraint programming languages. The wikipedia page on ILP has a good general introduction: https://en.wikipedia.org/wiki/Inductive_logic_programming The most recent survey article I know of is the following…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Tue, 22 Jan 2019 19:34:38 GMT

Ah, but this is the great thing about ILP, including the δILP system in the article- that the "models" are actually logical theories that can be directly inspected (by reading them!) and verified. So we don't have to wonder whether the model learned by the algorithm will generalise outside its training data- we can just check the learned theory. This is one of the theories learned by the δILP system for the less-than relation between images of numbers, taken from the pdf version of the paper [1]. Using Prolog as the notation (the transformation from the article is trivial): target…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Wed, 16 Jan 2019 21:56:21 GMT

Btw, if you have relational data and a few good people with strong computer science backgrounds rather than statisticians or mathematicians, have a look at Inductive Logic Programming. ILP is a set of machine learning techniques that learn logic programs from logic programs. The sample efficiency is on a class of its own and it generalises robustly from very little data[1]. I study ILP algorithms for my PhD. My research group has recently developed a new technique, Meta Interpretive Learning. Its canonical implementation is Metagol: https://github.com/metagol/metagol Please feel free to email…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by omeze

Sat, 01 Dec 2018 19:06:17 GMT

Its a long article but he’s advocating for a hybrid approach to deal with problems that DNNs encounter today which he believes are fundamental to DNNs (reasoning, causality). An example of such an approach taken with success is given at the end: https://arxiv.org/pdf/1711.04574.pdf

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Fri, 24 Aug 2018 18:58:37 GMT

Actually, I wasn't thinking of any specific methods, but now that you mentioned it, Inductive Logic Progamming (the subject of my PhD - see my comment about being biased) is a fine example. For a slightly more impartial opinion, here's a DeepMind paper that performs neural ILP: https://deepmind.com/blog/learning-explanatory-rules-noisy-d... The authors begin by extolling the virtues of ILP, including its generalisation abilities, as follows: Second, ILP systems tend to be impressively data-efficient, able to generalise well from a small handful of examples. [1] You can find more references to…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — mentioned by YeGoblynQueenne

Fri, 15 Jun 2018 09:37:25 GMT

Has there been a misunderstanding of my comment? I didn't mean to say that I expect "AI to go away". For one thing, I certainly do not equate all of AI to deep learning, or even machine learning. My point is that deep learning has generated a lot of excitement in the last few years, but as we're hitting inevitable diminishing returns the excitement is dying down and people are starting to think about how to move beyond current techniques. As usual, the last to catch on to the reversal of the old trend, or the new trend if you prefer, are the people who pay the money- industry investors or the…

Paper and discussion history. A mention is not an endorsement.

Learning Explanatory Rules from Noisy Data — HN submission

Fri, 16 Feb 2018 19:12:18 GMT

Learning Explanatory Rules from Noisy Data [with Neural Networks]

4 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Learning Explanatory Rules from Noisy Data — HN submission

Wed, 15 Nov 2017 16:21:56 GMT

Learning Explanatory Rules from Noisy Data [pdf]

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.