Evaluating Simple Fully Automated Heuristics for Adaptive Constraint Propagation

Paparrizou, Anastasia; Stergiou, Kostas

doi:10.1109/ictai.2012.123

Cited by 11 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pruning can be weaker than the full version, but the computational effort can be significantly reduced. In Stergiou (2008) and Paparrizou and Stergiou (2012), heuristics allow the solver to dynamically select AC or a stronger level of consistency (maxRPC) during the search depending on the variable/constraint. The approaches are based on the clustering effect.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive filtering strategy for numerical constraint satisfaction problems

Araya

Soto

Crawford

2015

Expert Systems with Applications

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

“…In order to offer a more interesting comparison, we adapted the finite-domain oriented mechanism proposed in Stergiou (2008Stergiou ( , 2012 to interval-based contractors.…”

Section: A Stergiou-based Adaptive Contractionmentioning

confidence: 99%

Adaptive filtering strategy for numerical constraint satisfaction problems

Araya

Soto

Crawford

2015

Expert Systems with Applications

View full text Add to dashboard Cite

“…However, its performance is unpredictable and can differ widely on similar instances. Further, maintaining a given consistency property during search has become a common practice [13,22,30,42,43] and new strategies for dynamically switching between consistency algorithms are being investigated [1][2][3]9,15,16,18,25,29,41,46,48]. While consistency algorithms can dramatically reduce the size of the search space, their impact on the CPU cost of search can vary widely.…”

Section: Introductionmentioning

confidence: 99%

Visualizations to Summarize Search Behavior

Howell

Choueiry

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

In this paper, we argue that metrics that assess the performance of backtrack search for solving a Constraint Satisfaction Problem should not be visualized and examined only at the end of search, but their evolution should be tracked throughout the search process in order to provide a more complete picture of the behavior of search. We describe a process that organizes search history by automatically recognizing qualitatively significant changes in the metrics that assess search performance. To this end, we introduce a criterion for quantifying change between two time instants and a summarization technique for organizing the history of search at controllable levels of abstraction. We validate our approach in the context of two algorithms for enforcing consistency: one that is activated by a surge of backtracking and the second that modifies the structure of the constraint graph. We also introduce a new visualization for exposing the behavior of variable ordering heuristics and validate its usefulness both as a standalone tool and when displayed alongside search history.

show abstract

“…Choosing the right level of local consistency for solving a problem requires finding a good trade-o↵ between the ability of this local consistency to remove inconsistent values, and the cost of the algorithm that enforces it. The works of [Ste08] and [PS12] suggest to take advantage of the power of strong propagation algorithms to reduce the search space while avoiding the high cost of maintaining them in the whole network. These methods result in a heuristic approach based on the monitoring of propagation events to dynamically adapt the level of local consistency (arc consistency or max restricted path consistency) to individual constraints.…”

Section: Introductionmentioning

confidence: 99%

Adapting Consistency in Constraint Solving

Balafrej

Bessière

Paparrizou

et al. 2016

Data Mining and Constraint Programming

View full text Add to dashboard Cite

State-of-the-art constraint solvers uniformly maintain the same level of local consistency (usually arc consistency) on all the instances. We propose two approaches to adjust the level of consistency depending on the instance and on which part of the instance we propagate. The first approach, parameterized local consistency, uses as parameter the stability of values, which is a feature computed by arc consistency algorithms during their execution. Parameterized local consistencies choose to enforce arc consistency or a higher level of local consistency to a value depending on whether the stability of the value is above or below a given threshold. In the adaptive version, the parameter is dynamically adapted during search, and so is the level of local consistency. In the second approach, we focus on partition-one-AC, a singleton-based consistency. We propose adaptive variants of partition-one-AC that do not necessarily run until having proved the fixpoint. The pruning can be weaker than the full version, but the computational e↵ort can be significantly reduced. Our experiments show that adaptive parameterized maxRPC and adaptive partition-one-AC can obtain significant speed-ups over arc consistency and over the full versions of maxRPC and partition-one-AC. ⇤ The results contained in this chapter have been presented in [BBCB13] and [BBBT14]. This work has been funded by the EU project ICON (FP7-284715).

show abstract

Evaluating Simple Fully Automated Heuristics for Adaptive Constraint Propagation

Cited by 11 publications

References 13 publications

Adaptive filtering strategy for numerical constraint satisfaction problems

Adaptive filtering strategy for numerical constraint satisfaction problems

Visualizations to Summarize Search Behavior

Adapting Consistency in Constraint Solving

Contact Info

Product

Resources

About