Download PDFOpen PDF in browser

Lookahead-Based SMT Solving

17 pagesPublished: October 23, 2018

Abstract

The lookahead approach for binary-tree-based search in constraint solving favors branching that provide the lowest upper bound for the remaining search space. The approach has recently been applied in instance partitioning in divide-and-conquer-based parallelization, but in general its connection to modern, clause-learning solvers is poorly understood. We show two ways of combining lookahead approach with a modern DPLL(T)-based SMT solver fully profiting from theory propagation, clause learning, and restarts. Our thoroughly tested prototype implementation is surprisingly efficient as an independent SMT solver on certain instances, in particular when applied to a non-convex theory, where the lookahead-based implementation solves 40% more unsatisfiable instances compared to the standard implementation.

Keyphrases: clause learning, lookahead heuristic, smt solving

In: Gilles Barthe, Geoff Sutcliffe and Margus Veanes (editors). LPAR-22. 22nd International Conference on Logic for Programming, Artificial Intelligence and Reasoning, vol 57, pages 418-434.

BibTeX entry
@inproceedings{LPAR-22:Lookahead_Based_SMT_Solving,
  author    = {Antti Hyvärinen and Matteo Marescotti and Parvin Sadigova and Hana Chockler and Natasha Sharygina},
  title     = {Lookahead-Based SMT Solving},
  booktitle = {LPAR-22. 22nd International Conference on Logic for Programming, Artificial Intelligence and Reasoning},
  editor    = {Gilles Barthe and Geoff Sutcliffe and Margus Veanes},
  series    = {EPiC Series in Computing},
  volume    = {57},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {/publications/paper/XVm6},
  doi       = {10.29007/gzzf},
  pages     = {418-434},
  year      = {2018}}
Download PDFOpen PDF in browser