DeepAI AI Chat
Log In Sign Up

DASH: Dynamic Approach for Switching Heuristics

by   Giovanni Di Liberto, et al.
Università di Padova
Monash University
University College Cork

Complete tree search is a highly effective method for tackling MIP problems, and over the years, a plethora of branching heuristics have been introduced to further refine the technique for varying problems. Recently, portfolio algorithms have taken the process a step further, trying to predict the best heuristic for each instance at hand. However, the motivation behind algorithm selection can be taken further still, and used to dynamically choose the most appropriate algorithm for each encountered subproblem. In this paper we identify a feature space that captures both the evolution of the problem in the branching tree and the similarity among subproblems of instances from the same MIP models. We show how to exploit these features to decide the best time to switch the branching heuristic and then show how such a system can be trained efficiently. Experiments on a highly heterogeneous collection of MIP instances show significant gains over the pure algorithm selection approach that for a given instance uses only a single heuristic throughout the search.


page 1

page 2

page 3

page 4


Deep Learning Assisted Heuristic Tree Search for the Container Pre-marshalling Problem

One of the key challenges for operations researchers solving real-world ...

Learning Heuristic Selection with Dynamic Algorithm Configuration

A key challenge in satisfying planning is to use multiple heuristics wit...

Rational Deployment of CSP Heuristics

Heuristics are crucial tools in decreasing search effort in varied field...

Mini-Bucket Heuristics for Improved Search

The paper is a second in a series of two papers evaluating the power of ...

Multiple-Goal Heuristic Search

This paper presents a new framework for anytime heuristic search where t...

Concurrent Cube-and-Conquer

Recent work introduced the cube-and-conquer technique to solve hard SAT ...