scholarly journals Constraint Programming for an Efficient and Flexible Block Modeling Solver

2020 ◽  
Vol 34 (09) ◽  
pp. 13685-13688
Author(s):  
Alex Lucía Mattenet ◽  
Ian Davidson ◽  
Siegfried Nijssen ◽  
Pierre Schaus
Keyword(s):  
Constraint Programming ◽  
Search Tree ◽  
Combinatorial Problems ◽  
Block Modeling ◽  
Clustering Problem ◽  
Constraint Language ◽  
Spatio Temporal ◽  
Feasible Solutions ◽  
High Level ◽  
Temporal Data Analysis

Constraint Programming (CP) is a powerful paradigm for solving combinatorial problems. In CP, the user creates a model by declaring variables with their domains and expresses the constraints that need to be satisfied in any solution. The solver is then in charge of finding feasible solutions—a value in the domain of each variable that satisfies all the constraints. The discovery of solutions is done by exploring a search tree that is pruned by the constraints in charge of removing impossible values. The CP framework has the advantage of exposing a rich high-level declarative constraint language for modeling, as well as efficient purpose-specific filtering algorithms that can be reused in many problems. In this work, we harness this flexibility and efficiency for the Block Modeling problem. It is a variant of the graph clustering problem that has been used extensively in many domains including social science, spatio-temporal data analysis and even medical imaging. We present a new approach based on constraint programming, allowing discrete optimization of block modeling in a manner that is not only scalable, but also allows the easy incorporation of constraints. We introduce a new constraint filtering algorithm that outperforms earlier approaches. We show its use in the analysis of real datasets.

scholarly journals Constraint Problem Specification as Compression

10.29007/7ths ◽  
2018 ◽  
Author(s):  
Steve Prestwich ◽  
S. Armagan Tarim ◽  
Roberto Rossi
Keyword(s):  
Logic Programming ◽  
Data Compression ◽  
Constraint Programming ◽  
Auxiliary Problem ◽  
Combinatorial Problems ◽  
Specification Languages ◽  
Constraint Problem ◽  
High Level ◽  
Natural Way ◽  
Comparable Size

Constraint Programming is a powerful and expressive framework for modelling and solving combinatorial problems. It is nevertheless not always easy to use, which has led to the development of high-level specification languages. We show that Constraint Logic Programming can be used as a meta-language to describe itself more compactly at a higher level of abstraction. This can produce problem descriptions of comparable size to those in existing specification languages, via techniques similar to those used in data compression. An advantage over existing specification languages is that, for a problem whose specification requires the solution of an auxiliary problem, a single specification can unify the two problems. Moreover, using a symbolic representation of domain values leads to a natural way of modelling channelling constraints.

scholarly journals Symmetry breaking in a new stable model search method

10.29007/1l5r ◽  
2018 ◽  
Author(s):  
Tarek Khaled ◽  
Belaid Benhamou
Keyword(s):  
Knowledge Representation ◽  
Symmetry Breaking ◽  
Constraint Programming ◽  
Answer Set Programming ◽  
Search Tree ◽  
Search Method ◽  
Combinatorial Problems ◽  
Knowledge Representation And Reasoning ◽  
Stable Model ◽  
Model Search

In this work, we investigate the inclusion of symmetry breaking in the answer set programming (ASP) framework. The notion of symmetry is widely studied in various domains. Particularly, in the field of constraint programming, where symmetry breaking made a significant improvement in the performances of many constraint solvers. Usually, combinatorial problems contain a lot of symmetries that could render their resolution difficult for the solvers that do not consider them. Indeed, these symmetries guide the solvers in the useless exploration of symmetric and redundant branches of the search tree. The ASP framework is well-known in knowledge representation and reasoning. How- ever, only few works on symmetry in ASP exist. We propose in this paper a new ASP solver based on a novel semantics that we enhance by symmetry breaking. This method with symmetry elimination is implemented and used for the resolution of a large variety of combinatorial problems. The obtained results are very promising and showcase an advantage when using our method in comparison to other known ASP methods.

scholarly journals From Support Propagation to Belief Propagation in Constraint Programming

2019 ◽  
Vol 66 ◽  
Author(s):  
Gilles Pesant
Keyword(s):  
Constraint Programming ◽  
Belief Propagation ◽  
Combinatorial Problems ◽  
Global Constraints ◽  
Inference Algorithms ◽  
Shared Variables ◽  
Shared Information ◽  
Individual Variables ◽  
High Level ◽  
Search Guidance

The distinctive driving force of constraint programming to solve combinatorial problems has been a privileged access to problem structure through the high-level models it uses. From that exposed structure in the form of so-called global constraints, powerful inference algorithms have shared information between constraints by propagating it through shared variables’ domains, traditionally by removing unsupported values. This paper investigates a richer propagation medium made possible by recent work on counting solutions inside constraints. Beliefs about individual variable-value assignments are exchanged between contraints and iteratively adjusted. It generalizes standard support propagation and aims to converge to the true marginal distributions of the solutions over individual variables. Its advantage over standard belief propagation is that the higher-level models featuring large-arity (global) constraints do not tend to create as many cycles, which are known to be problematic for convergence. The necessary architectural changes to a constraint programming solver are described and an empirical study of the proposal is conducted on its implementation. We find that it provides close approximations to the true marginals and that it significantly improves search guidance.

scholarly journals From Support Propagation to Belief Propagation in Constraint Programming (Extended Abstract)

2020 ◽  
Author(s):  
Gilles Pesant
Keyword(s):  
Constraint Programming ◽  
Driving Force ◽  
Belief Propagation ◽  
Combinatorial Problems ◽  
Problem Structure ◽  
Privileged Access ◽  
Propagation Medium ◽  
High Level ◽  
Search Guidance ◽  
Counting Solutions

The distinctive driving force of constraint programming (CP) to solve combinatorial problems has been a privileged access to problem structure through the high-level models it uses. We investigate a richer propagation medium for CP made possible by recent work on counting solutions inside constraints. Beliefs about individual variable-value assignments are exchanged between contraints and iteratively adjusted. Its advantage over standard belief propagation is that the higher-level models do not tend to create as many cycles, which are known to be problematic for convergence. We find that it significantly improves search guidance.

scholarly journals Generic Constraint-based Block Modeling using Constraint Programming

2021 ◽  
Vol 70 ◽  
pp. 597-630
Author(s):  
Alex Mattenet ◽  
Ian Davidson ◽  
Siegfried Nijssen ◽  
Pierre Schaus
Keyword(s):  
Discrete Optimization ◽  
Constraint Programming ◽  
Minimum Description Length ◽  
Mixed Integer ◽  
Neighborhood Search ◽  
Integer Programming Problem ◽  
Subsequent Work ◽  
Block Modeling ◽  
Temporal Data Analysis ◽  
Mixed Integer Programming Problem

Block modeling has been used extensively in many domains including social science, spatial temporal data analysis and even medical imaging. Original formulations of the problem modeled it as a mixed integer programming problem, but were not scalable. Subsequent work relaxed the discrete optimization requirement, and showed that adding constraints is not straightforward in existing approaches. In this work, we present a new approach based on constraint programming, allowing discrete optimization of block modeling in a manner that is not only scalable, but also allows the easy incorporation of constraints. We introduce a new constraint filtering algorithm that outperforms earlier approaches, in both constrained and unconstrained settings, for an exhaustive search and for a type of local search called Large Neighborhood Search. We show its use in the analysis of real datasets. Finally, we show an application of the CP framework for model selection using the Minimum Description Length principle.

Damage Detection in Flexible Propeller Beam Structures by Exploiting Impact-Induced Coupled Acceleration Signals

2011 ◽  
Author(s):  
Ioannis T. Georgiou
Keyword(s):  
Data Driven ◽  
Distributed Information ◽  
Modal Expansion ◽  
Shape Difference ◽  
Spatio Temporal ◽  
Temporal Sample ◽  
High Level ◽  
Proper Orthogonal ◽  
Acceleration Signals

A local damage at the tip of a composite propeller is diagnosed by properly comparing its impact-induced free coupled dynamics to that of a pristine wooden propeller of the same size and shape. This is accomplished by creating indirectly via collocated measurements distributed information for the coupled acceleration field of the propellers. The powerful data-driven modal expansion analysis delivered by the Proper Orthogonal Decomposition (POD) Transform reveals that ensembles of impact-induced collocated coupled experimental acceleration signals are underlined by a high level of spatio-temporal coherence. Thus they furnish a valuable spatio-temporal sample of coupled response induced by a point impulse. In view of this fact, a tri-axial sensor was placed on the propeller hub to collect collocated coupled acceleration signals induced via modal hammer nondestructive impacts and thus obtained a reduced order characterization of the coupled free dynamics. This experimental data-driven analysis reveals that the in-plane unit components of the POD modes for both propellers have similar shapes-nearly identical. For the damaged propeller this POD shape-difference is quite pronounced. The shapes of the POD modes are used to compute indices of difference reflecting directly damage. At the first POD energy level, the shape-difference indices of the damaged composite propeller are quite larger than those of the pristine wooden propeller.

scholarly journals A methodological roadmap to quantify animal-vectored spatial ecosystem subsidies

2020 ◽  
Author(s):  
Diego Ellis-Soto ◽  
Kristy M. Ferraro ◽  
Matteo Rizzuto ◽  
Emily Briggs ◽  
Julia D. Monk ◽  
...  
Keyword(s):  
Animal Movement ◽  
Open Systems ◽  
Nutrient Transport ◽  
Ecosystem Dynamics ◽  
Ecosystem Impacts ◽  
Practical Applications ◽  
Ecosystem Theory ◽  
Spatio Temporal ◽  
High Level ◽  
Tracking Devices

Ecosystems are open systems connected through spatial flows of energy, matter, and nutrients. Predicting and managing ecosystem interdependence requires a rigorous quantitative understanding of the drivers and vectors that connect ecosystems across spatio-temporal scales. Animals act as such vectors when they transport nutrients across landscapes in the form of excreta, egesta, and their own bodies. Here, we introduce a methodological roadmap that combines movement, foraging, and ecosystem ecology to study the effects of animal-vectored nutrient transport on meta-ecosystems. The meta-ecosystem concept — the notion that ecosystems are connected in space and time by flows of energy, matter, and organisms across boundaries — provides a theoretical framework on which to base our understanding of animal-vectored nutrient transport. However, partly due to its high level of abstraction, there are few empirical tests of meta-ecosystem theory, and while we may label animals as important mediators of ecosystem services, we lack predictive inference of their relative roles and impacts on diverse ecosystems. Recently developed technologies and methods — tracking devices, mechanistic movement models, diet reconstruction techniques and remote sensing — have the potential to facilitate the quantification of animal-vectored nutrient flows and increase the predictive power of meta-ecosystem theory. Understanding the mechanisms by which animals shape ecosystem dynamics may be important for ongoing conservation, rewilding, and restoration initiatives around the world, and for more accurate models of ecosystem nutrient budgets. We provide conceptual examples that show how our proposed integration of methodologies could help investigate ecosystem impacts of animal movement. We conclude by describing practical applications to understanding cross-ecosystem contributions of animals on the move.

scholarly journals SPATIO-TEMPORAL DATA ANALYSIS WITH NON-LINEAR FILTERS: BRAIN MAPPING WITH fMRI DATA

2011 ◽  
Vol 19 (3) ◽  
pp. 189
Author(s):  
Karsten Rodenacker ◽  
Klaus Hahn ◽  
Gerhard Winkler ◽  
Dorothea P Auer
Keyword(s):  
Signal To Noise Ratio ◽  
Three Dimensional ◽  
Digital Data ◽  
Linear Filters ◽  
Brain Functions ◽  
Non Linear ◽  
Long Time ◽  
Spatio Temporal ◽  
Sensory Activation ◽  
Temporal Data Analysis

Spatio-temporal digital data from fMRI (functional Magnetic Resonance Imaging) are used to analyse and to model brain activation. To map brain functions, a well-defined sensory activation is offered to a test person and the hemodynamic response to neuronal activity is studied. This so-called BOLD effect in fMRI is typically small and characterised by a very low signal to noise ratio. Hence the activation is repeated and the three dimensional signal (multi-slice 2D) is gathered during relatively long time ranges (3-5 min). From the noisy and distorted spatio-temporal signal the expected response has to be filtered out. Presented methods of spatio-temporal signal processing base on non-linear concepts of data reconstruction and filters of mathematical morphology (e.g. alternating sequential morphological filters). Filters applied are compared by classifications of activations.

Sign in / Sign up

Export Citation Format

Share Document