Learning union of integer hypercubes with queries (with applications to monadic decomposition)

DOI10.1007/978-3-030-81688-9_12zbMATH Open1493.68170arXiv2105.13071OpenAlexW3186357476MaRDI QIDQ832265FDOQ832265

Oliver Markgraf, Daniel Stan, Anthony Widjaja Lin

Publication date: 25 March 2022

Abstract: We study the problem of learning a finite union of integer (axis-aligned) hypercubes over the d-dimensional integer lattice, i.e., whose edges are parallel to the coordinate axes. This is a natural generalization of the classic problem in the computational learning theory of learning rectangles. We provide a learning algorithm with access to a minimally adequate teacher (i.e. membership and equivalence oracles) that solves this problem in polynomial-time, for any fixed dimension d. Over a non-fixed dimension, the problem subsumes the problem of learning DNF boolean formulas, a central open problem in the field. We have also provided extensions to handle infinite hypercubes in the union, as well as showing how subset queries could improve the performance of the learning algorithm in practice. Our problem has a natural application to the problem of monadic decomposition of quantifier-free integer linear arithmetic formulas, which has been actively studied in recent years. In particular, a finite union of integer hypercubes correspond to a finite disjunction of monadic predicates over integer linear arithmetic (without modulo constraints). Our experiments suggest that our learning algorithms substantially outperform the existing algorithms.

Full work available at URL: https://arxiv.org/abs/2105.13071

Recommendations

• Learning unions of high-dimensional boxes over the reals
• Learning Unions of ω(1)-Dimensional Rectangles
• Learning boxes in high dimension
• Exact Learning of Discretized Geometric Concepts
• Learning unions of \(\omega(1)\)-dimensional rectangles

Mathematics Subject Classification ID

Computational learning theory (68Q32) Logic in computer science (03B70)

Cites Work

• Learning regular sets from queries and counterexamples
• Queries and concept learning
• Learning Minimal Separating DFA’s for Compositional Verification
• When won't membership queries help?
• Characterizing PAC-learnability of semilinear sets
• Decision procedures. An algorithmic point of view
• On the number of prime implicants
• Monadic Decomposition
• The learnability of unions of two rectangles in the two-dimensional discretized space
• Learning semilinear sets from examples and via queries
• Algorithms and lower bounds for on-line learning of geometrical concepts
• Parameterized synthesis with safety properties
• Monadic decomposition in integer linear arithmetic
• The power of symbolic automata and transducers
• Decision problems among the main subfamilies of rational relations
• Liveness of Randomised Parameterised Systems under Arbitrary Schedulers
• Title not available (Why is that?)
• Variable independence for first-order definable constraints
• A constraint-based approach to solving games on infinite graphs
• An optimal algorithm for proper learning of unions of two rectangles with queries

Cited In (1)

• Finding the Rare Cube

Uses Software

• z3

This page was built for publication: Learning union of integer hypercubes with queries (with applications to monadic decomposition)

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q832265)