Subproblem ordering heuristics for AND/OR best-first search

Best-first search can be regarded as anytime scheme for producing lower bounds on the optimal solution, a characteristic that is mostly overlooked. We explore this topic in the context of AND/OR best-first search, guided by the MBE heuristic, when solving graphical models. In that context, the impac...

Full description

Bibliographic Details
Authors: Lam, William, Kask, Kalev, Larrosa Bondia, Francisco Javier|||0000-0002-8322-0505, Dechter, Rina
Format: article
Publication Date:2017
Country:España
Institution:Universitat Politècnica de Catalunya (UPC)
Repository:UPCommons. Portal del coneixement obert de la UPC
Language:English
OAI Identifier:oai:upcommons.upc.edu:2117/111987
Online Access:https://hdl.handle.net/2117/111987
https://dx.doi.org/10.1016/j.jcss.2017.10.003
Access Level:Open access
Keyword:Graphic methods
Combinatorial optimization
Artificial intelligence
Graphical models
Heuristic search
Mètodes gràfics
Optimització combinatòria
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
Description
Summary:Best-first search can be regarded as anytime scheme for producing lower bounds on the optimal solution, a characteristic that is mostly overlooked. We explore this topic in the context of AND/OR best-first search, guided by the MBE heuristic, when solving graphical models. In that context, the impact of the secondary heuristic for subproblem ordering may be significant, especially in the anytime context. Indeed, our paper illustrates this, showing that a new concept of bucket errors can advise in providing effective subproblem orderings in AND/OR search for both exact and anytime solutions.