Learnability and automatizability

Misha Alekhnovich, Mark Braverman, Vitaly Feldman, Adam R. Klivans, Toniann Pitassi

Research output: Contribution to journalConference articlepeer-review

28 Scopus citations

Abstract

We consider the complexity of properly learning concept classes, i.e. when the learner must output a hypothesis of the same form as the unknown concept. We present the following new upper and lower bounds on well-known concept classes: We show that unless NP = RP, there is no polynomial-time PAC learning algorithm for DNF formulae where the hypothesis is an OR-of-thresholds. Note that as special cases, we show that neither DNF nor OR-of-thresholds are properly leamable unless N P = RP. Previous hardness results have required strong restrictions on the size of the output DNF formula. We also prove that it is HP-hard to learn the intersection of l ≥2 halfspaces by the intersection of k halfspaces for any constant k ≥ 0. Previous work held for the case when k = l. Assuming that NP ⊈ DTIME(2n∈) for a certain constant ∈. < 1 we show that it is not possible to learn size s decision trees by size sk decision trees for any k ≥ 0. Previous hardness results for learning decision trees held for k ≤ 2. We present the first non-trivial upper bounds on properly learning DNF formulae and decision trees. In particular we show how to learn size s DNF by DNF in time 2Ō(√n log s) and how to learn size s decision trees by decision trees in time n O(log s). The hardness results for DNF formulae and intersections of halfspaces are obtained via specialized graph products for amplifying the hardness of approximating the chromatic number as well as applying recent work on the hardness of approximate hypergraph coloring. The hardness results for decision trees, as well as the new upper bounds, are obtained by developing a connection between automatizability in proof complexity and learnability, which may have other applications.

Original languageEnglish (US)
Pages (from-to)621-630
Number of pages10
JournalProceedings - Annual IEEE Symposium on Foundations of Computer Science, FOCS
StatePublished - 2004
Externally publishedYes
EventProceedings - 45th Annual IEEE Symposium on Foundations of Computer Science, FOCS 2004 - Rome, Italy
Duration: Oct 17 2004Oct 19 2004

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Fingerprint Dive into the research topics of 'Learnability and automatizability'. Together they form a unique fingerprint.

Cite this