Top Program Construction and Reduction for polynomial time Meta-Interpretive Learning · HackerPapers discussions

Top Program Construction and Reduction for polynomial time Meta-Interpretive Learning — mentioned by YeGoblynQueenne

Mon, 24 Jul 2023 19:56:34 GMT

Of course. It's published as an open-access paper here: https://link.springer.com/article/10.1007/s10994-020-05945-w But I recommend the arxiv version where Springer couldn't mess up my LaTex formatting: https://arxiv.org/abs/2101.05050 (Not that my LaTex formatting is anything to write home about!). To clarify, the algorithm described in our paper doesn't reduce the size of the search space for logic programs- it sidesteps it. Note I'm the corresponding author in the paper and my email is in the Springer version (top of the page). I'm always happy to answer questions about my work.

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

Top Program Construction and Reduction for polynomial time Meta-Interpretive Learning — mentioned by YeGoblynQueenne

Tue, 20 Apr 2021 17:42:13 GMT

It is true that these were problems with ILP. But, have a look at the Top Program Construction paper: https://arxiv.org/abs/2101.05050 The experiments section shows one example of learning a program with a few thousand clauses and one example of learning with classification noise (i.e. misclassified examples), all in polynomial time and without "magic parameters" to overcome noise. P.S. Er, actually I'm the corresponding author of that paper so eh, take my comment with a grain of salt as usual when researchers promote their research...

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