Balafoutis, Thanasis; Balaska, Vasiliki; Gasteratos, Antonios TraDance: A dataset for assessing the performance of Greek folk dances recognition models Conference In Proceedings of the IEEE/IST International Conference on Imaging Systems & Techniques, Copenhagen, Denmark , 2023. Tragiannis, Konstantinos; Balafoutis, Thanasis; Balaska, Vasiliki; Gasteratos, Antonios Skeleton-Based Recognition of Traditional Greek Dance Steps using Machine Learning Algorithms Conference In Proceedings of the IEEE/IST International Conference on Imaging Systems & Techniques, Copenhagen, Denmark, 2023. Vogt, Joachim; Marghitu, Octav; Blagau, Adrian; Pick, Leonie; Stachlys, Nele; Buchert, Stephan; Sarris, Theodoros; Tourgaidis, Stelios; Balafoutis, Thanasis; Baloukidis, Dimitrios; Pirnaris, Panagiotis Daedalus Ionospheric Profile Continuation Journal Article In: Geoscientific Instrumentation, Methods and Data Systems, in review, 2022. Sarris, Theodoros E.; Balafoutis, Thanasis; Kottaras, George A Greek QB50 nano-satellite for Upper Atmosphere Studies Miscellaneous Hipparchos. The Hellenic Astronomical Society Newsletter, ISSN: 1790-9252, Volume 2, Issue 12, pages 11-17, 2015, ISSN: 1790-9252. Balafoutis, Thanasis The Importance of Variable Ordering in Constraint Satisfaction Problems Journal Article In: Advanced Computational Technologies, Romanian Academy Publishing House, vol. 322, pp. 206-216, 2012. Balafoutis, Thanasis; Paparrizou, Anastasia; Stergiou, Kostas; Walsh, Toby New Algorithms for max Restricted Path Consistency Journal Article In: Constraints, Springer, vol. 16, no. 4, pp. 372-406, 2011. Balafoutis, Thanasis Adaptive Strategies for Solving Constraint Satisfaction Problems PhD Thesis Artificial Intelligence Laboratory, Department of Information and Communication Systems Engineering, University of the Aegean, 2011. Balafoutis, Thanasis; Paparrizou, Anastasia; Stergiou, Kostas Experimental Evaluation of Branching Schemes for the CSP Workshop In Proceedings of the TRICS workshop at 6th International Conference on Principles and Practice of Constraint Programming (CP 2010), pages 1–12, St. Andrews, Scotland, 2010. Balafoutis, Thanasis; Paparrizou, Anastasia; Stergiou, Kostas; Walsh, Toby Improving the performance of maxRPC Conference In Proceedings of the 16th International Conference on Principles and Practice of Constraint Programming (CP 2010), LNCS, Vo 6308, pp. 69-83, St Andrews, Scotland, vol. 6308, 2010. Balafoutis, Thanasis; Stergiou, Kostas Evaluating and Improving Modern Variable and Revision Ordering Strategies in CSPs Journal Article In: Fundamenta Informaticae, IOS Press, vol. 102, iss. 3-4, pp. 229-261, 2010. Balafoutis, Thanasis; Stergiou, Kostas Adaptive branching for constraint satisfaction problems Conference In Proceedings of the 19th European Conference on Artificial Intelligence (ECAI 2010), pages 855–860, Lisbon, Portugal, 2010. Balafoutis, Thanasis; Stergiou, Kostas Conflict directed variable selection strategies for constraint satisfaction problems Conference In Proceedings of the 6th Hellenic Conference on Artificial Intelligence (SETN), pages 29–38, Athens, 2010. Balafoutis, Thanasis; Panagiotis Pavlos, Eleftherios Pavlos Boosting Wikis with Mind Maps in Collaborative Learning Environments Conference In Proceedings of the 6th National and International HSSS Conference "Systemic Approaches in Social Structures", Mytilene, 2010. Tsinakos, Avgoustos; Balafoutis, Thanasis A Comparative Survey On Mind Mapping Tools Journal Article In: Turkish Online Journal of Distance Education-TOJDE, vol. 10, no. 3, pp. 55-67, 2009, ISSN: 1302-6488. Balafoutis, Thanasis; Stergiou, Kostas Experimental Evaluation of Modern Variable Selection Strategies in Constraint Satisfaction Problems Workshop In Proceedings of the 15th RCRA workshop on Experimental Evaluation of Algorithms for Solving Problems with Combinatorial Explosion, Udine, Italy, 2008. Balafoutis, Thanasis; Stergiou, Kostas Exploiting constraint weights for revision ordering in Arc Consistency Algorithms Workshop In Proceedings of the European Conference on Artificial Intelligence Workshop on Modeling and Solving Problems with Constraints, Patras, Greece, 2008. Balafoutis, Thanasis; Stergiou, Kostas On Conflict-driven variable ordering heuristics Workshop In Proceedings of 13th Annual ERCIM International Workshop on Constraint Solving and Constraint Logic Programming, CSCLP-08, Rome, Italy, 2008. Balafoutis, Thanasis; Stergiou, Kostas Algorithms for stochastic CSPs Conference In Proceedings of the 12th International Conference on Principles and Practices of Constraint Programming, LNCS 4204, pages 44–58, Nantes, France, 2006. Balafoutis, Thanasis Algorithms for Stochastic Constraint Satisfaction Problems Masters Thesis Artificial Intelligence Laboratory, Department of Information and Communication Systems Engineering, University of the Aegean, 2006.@conference{Balafoutis2023,
title = {TraDance: A dataset for assessing the performance of Greek folk dances recognition models},
author = {Thanasis Balafoutis and Vasiliki Balaska and Antonios Gasteratos},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2023/11/traDance-1.pdf, IST2023.pdf},
year = {2023},
date = {2023-10-17},
urldate = {2023-10-17},
booktitle = {In Proceedings of the IEEE/IST International Conference on Imaging Systems & Techniques, Copenhagen, Denmark },
abstract = {In computer vision, datasets and benchmarks are widely used to compare algorithms and boost scientific progress. Especially in the human action recognition research field, extracting dance poses from video sequences for fragmentation and recognition of dancer movements is a challenging task, and new datasets are always important. This paper presents a new benchmark dataset and evaluation methodology for dancing analysis and step movements classification. We focus on traditional Greek dances, which are characterized by the repetition of simple steps throughout the duration of the dance track. The dataset, named TraDance, contains 1.4 hours of 3D dance motion videos from 5 different experienced dancers and 3494 annotated dance steps, covering 5 traditional dance genres with known RGB-D camera settings. In addition, to evaluate the quality of TraDance, we provide extensive experiments using existing machine learning algorithms that are commonly used in computer vision. The results show that TraDance is feasible for research-purpose studies and reliable. The dataset can be found online at: https://robotics.pme.duth.gr/tradance.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@conference{Tragiannis2023,
title = {Skeleton-Based Recognition of Traditional Greek Dance Steps using Machine Learning Algorithms},
author = {Konstantinos Tragiannis and Thanasis Balafoutis and Vasiliki Balaska and Antonios Gasteratos},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2023/11/sbr.pdf, IST2023.pdf},
year = {2023},
date = {2023-10-16},
urldate = {2023-10-16},
booktitle = {In Proceedings of the IEEE/IST International Conference on Imaging Systems & Techniques, Copenhagen, Denmark},
abstract = {Classification of dance types and moves is a challenging application in the field of action recognition. Owing to its vast applicability and robustness regardless of the image background, skeleton-based recognition is increasingly attracting the attention of scholars in the robotics community. This paper presents a novel dataset of videos of five individuals performing five different types of traditional Greek dances, which can be utilized for that purpose. In addition, we create a benchmark for future work by performing a comparative study of different classifiers' performance on dance steps from this dataset. For that purpose, the dataset has been divided into steps, with each step consisting of a time series of skeleton data extracted from the videos. We use an assortment of classifiers based on different artificial neural network (NN) types, including convolutional and recurrent NNs as well as traditional ones such as Gaussian naive Bayes, decision trees, support vector machines, and $k$-nearest neighbors. For each classifier, we have obtained the classification accuracy first on steps from the same dance type and then on the entire dataset. We assess the classifiers' performance both on randomly split test sets and when performing cross-validation. Out of the classifiers used, the best performance is achieved by the CNN-based classifier. },
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@article{Vogt2022,
title = {Daedalus Ionospheric Profile Continuation},
author = {Joachim Vogt and Octav Marghitu and Adrian Blagau and Leonie Pick and Nele Stachlys and Stephan Buchert and Theodoros Sarris and Stelios Tourgaidis and Thanasis Balafoutis and Dimitrios Baloukidis and Panagiotis Pirnaris},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/gi-2022-12.pdf, GI2022.pdf},
year = {2022},
date = {2022-08-01},
urldate = {2022-08-01},
journal = {Geoscientific Instrumentation, Methods and Data Systems, in review},
abstract = {The Daedalus Ionospheric Profile Continuation (DIPCont) project is concerned with the question how in situ measurements in the lower thermosphere and ionosphere (LTI) can be extrapolated using parametric models of observables and derived variables. To reflect the pronounced change of temperature across the LTI, non-isothermal models for neutral density and also electron density are constructed from scale height profiles that increase linearly with altitude. Ensembles of model parameters are created by means of Monte Carlo simulations using synthetic measurements based on model predictions and relative uncertainties as specified in the Daedalus Report for Assessment. The parameter ensembles give rise to ensembles of model altitude profiles for LTI variables of interest. Extrapolation quality is quantified by statistics derived from the altitude profile ensembles. The vertical extent of meaningful profile continuation is captured by the concept of extrapolation horizons defined as the boundaries of regions where the deviations remain below a prescribed error threshold. The methodology allows for assessing how cost-critical elements of the Daedalus mission proposal such as perigee and apogee distances as major factors controling the necessary amount of propellant and radiation shielding, respectively, affect the accuracy of scientific inference in the LTI. First results are presented for dual-satellite measurements at different inter-spacecraft distances but also for the single-satellite case to compare the two basic mission scenarios under consideration. DIPCont models and procedures are implemented in a collection of Python modules and Jupyter notebooks supplementing this report.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
@misc{Sarris2015,
title = {A Greek QB50 nano-satellite for Upper Atmosphere Studies},
author = {Theodoros E. Sarris and Thanasis Balafoutis and George Kottaras},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/hipparchos2015.pdf, HIPPARCHOS2015.pdf},
issn = {1790-9252},
year = {2015},
date = {2015-06-01},
urldate = {2015-06-01},
booktitle = {Hipparchos. The Hellenic Astronomical Society Newsletter, ISSN: 1790-9252, Volume 2, Issue 12, pages 11-17},
volume = {2},
issue = {12},
pages = {11-17},
abstract = {The Laboratory of Electromagnetism and Space Research of the Democritus University of Thrace (DUTH/SRL) has been selected to join the QB50 European initiative for the launch of 50 nano-satellites in the upper atmosphere by January 2016. The aim is to investigate with multi-point measurements the transition region between the atmosphere and space. The 50 nano-satellites follow the CubeSat standard, where a CubeSat is a modular satellite of standardized dimensions, assembled using primarily commercial, off-the shelf components. This provides an excellent opportunity for the launch of a Greek miniaturized satellite that is entirely built by University students and engineers. Through the QB50 program a launch opportunity and part of the science payload are provided whereas the development of each CubeSat and the ground station for communications and operations are built by the host institution. In this paper we present the objectives of the QB50 mission and the status of development of the Greek QB50 CubeSat.},
howpublished = {Hipparchos. The Hellenic Astronomical Society Newsletter, ISSN: 1790-9252, Volume 2, Issue 12, pages 11-17},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
@article{Balafoutis2012,
title = {The Importance of Variable Ordering in Constraint Satisfaction Problems},
author = {Thanasis Balafoutis},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/actr2012.pdf, ACTR2012.pdf},
year = {2012},
date = {2012-02-01},
urldate = {2012-02-01},
journal = {Advanced Computational Technologies, Romanian Academy Publishing House},
volume = {322},
pages = {206-216},
abstract = {Constraint programming is a powerful technique for solving combinatorial search problems that draws on a wide range of methods from artificial intelligence and computer science. The basic idea in constraint programming is that the user states the constraints and a general purpose constraint solver is used to solve the resulting constraint satisfaction problem. A key factor that can dramatically reduce the search space during constraint solving is the criterion under which the variable to be instantiated next is selected. For this purpose numerous strategies have been proposed. Some of the best of such strategies exploit information about failures gathered throughout search and recorded in the form of constraint weights, while others measure the importance of variable assignments in reducing the search space. In this work we give an introduction on the constraint satisfaction problems. We have also collect and present the most recent and powerful variable ordering heuristics that have been proposed in the literature. Our intention is to provide a comprehensive view of the relative strengths and weaknesses of these heuristics. We also provide insight as to which heuristic is preferable as a general purpose variable ordering strategy.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
@article{Balafoutis2011,
title = {New Algorithms for max Restricted Path Consistency},
author = {Thanasis Balafoutis and Anastasia Paparrizou and Kostas Stergiou and Toby Walsh},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/maxRPC_journal.pdf, Constraints2011.pdf},
year = {2011},
date = {2011-10-01},
urldate = {2011-10-01},
journal = {Constraints, Springer},
volume = {16},
number = {4},
pages = {372-406},
publisher = {Springer},
abstract = {Max Restricted Path Consistency (maxRPC) is a local consistency for binary constraints that enforces a higher order of consistency than arc consistency. Despite the strong pruning that can be achieved, maxRPC is rarely used because existing maxRPC algorithms suffer from overheads and redundancies as they can repeatedly perform many constraint checks without triggering any value deletions. In this paper we propose and evaluate techniques that can boost the performance of maxRPC algorithms by eliminating many of these overheads and redundancies. These include the combined use of two data structures to avoid many redundant constraint checks, and the exploitation of residues to quickly verify the existence of supports. Based on these, we propose a number of closely related maxRPC algorithms. The first one, maxRPC3, has optimal O(end 3) time complexity, displays good performance when used stand-alone, but is expensive to apply during search. The second one, maxRPC3rm, has O(en2d4) time complexity, but a restricted version
with O(end4) complexity can be very efficient when used during search. The other algorithms are simple modifications of maxRPC3rm. All algorithms have O(ed) space complexity when used stand-alone. However, maxRPC3 has O(end) space complexity when used during search, while the others retain the O(ed) complexity. Experimental results demonstrate that the resulting methods constantly outperform previous algorithms for maxRPC, often by large margins, and constitute a viable alternative to arc consistency on some problem classes.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
with O(end4) complexity can be very efficient when used during search. The other algorithms are simple modifications of maxRPC3rm. All algorithms have O(ed) space complexity when used stand-alone. However, maxRPC3 has O(end) space complexity when used during search, while the others retain the O(ed) complexity. Experimental results demonstrate that the resulting methods constantly outperform previous algorithms for maxRPC, often by large margins, and constitute a viable alternative to arc consistency on some problem classes.@phdthesis{Balafoutis2011b,
title = {Adaptive Strategies for Solving Constraint Satisfaction Problems},
author = {Thanasis Balafoutis},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/11/Thesis.pdf, Thesis.pdf},
year = {2011},
date = {2011-06-01},
urldate = {2011-06-01},
school = {Artificial Intelligence Laboratory, Department of Information and Communication Systems Engineering, University of the Aegean},
abstract = {A major challenge in constraint programming is to develop efficient generic approaches to solve instances of the constraint satisfaction problem (CSP). In recent years, adaptive approaches for solving CSPs have attracted the interest of many researchers. General speaking, a strategy that uses the results of its own search experience to modify its subsequent behavior does adaptive search.
In this dissertation we explore adaptive strategies for backtracking search on various levels. First, we investigate adaptive search-guiding heuristics for ordering variables in CSPs. These adaptive heuristics learn and use information from every node explored in the search tree, whereas traditional static and dynamic heuristics only use information about the initial and current nodes. We then perform a wide empirical evaluation of the proposed variable ordering heuristics and compare them with the current state-of-the-art variable ordering strategies.
Concerning constraint propagation which is used as an inference mechanism in order to simplify a problem so as to make it easier to solve, we explore adaptive strategies for ordering the different revisions performed when enforcing arc consistency algorithms. Next, we propose adaptive branching heuristics for splitting the search tree. The application of these heuristics results in an adaptive branching scheme. Experiments with instantiations of the proposed generic heuristics confirm that search with adaptive branching outperforms search with a fixed branching scheme on a wide range of problem.
Finally, we propose a new a generic approach for branching where the variable’s domains are grouped into sets by using the scores assigned to values by a value ordering heuristic, and a clustering algorithm from machine learning. In general, this dissertation contributes to the design and implementation of adaptive and autonomous constraint solvers that have the ability to advantageously modify modelers decisions that typically in mainstream CP solvers are taken prior to search.},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
In this dissertation we explore adaptive strategies for backtracking search on various levels. First, we investigate adaptive search-guiding heuristics for ordering variables in CSPs. These adaptive heuristics learn and use information from every node explored in the search tree, whereas traditional static and dynamic heuristics only use information about the initial and current nodes. We then perform a wide empirical evaluation of the proposed variable ordering heuristics and compare them with the current state-of-the-art variable ordering strategies.
Concerning constraint propagation which is used as an inference mechanism in order to simplify a problem so as to make it easier to solve, we explore adaptive strategies for ordering the different revisions performed when enforcing arc consistency algorithms. Next, we propose adaptive branching heuristics for splitting the search tree. The application of these heuristics results in an adaptive branching scheme. Experiments with instantiations of the proposed generic heuristics confirm that search with adaptive branching outperforms search with a fixed branching scheme on a wide range of problem.
Finally, we propose a new a generic approach for branching where the variable’s domains are grouped into sets by using the scores assigned to values by a value ordering heuristic, and a clustering algorithm from machine learning. In general, this dissertation contributes to the design and implementation of adaptive and autonomous constraint solvers that have the ability to advantageously modify modelers decisions that typically in mainstream CP solvers are taken prior to search.@workshop{Balafoutis2010b,
title = {Experimental Evaluation of Branching Schemes for the CSP},
author = {Thanasis Balafoutis and Anastasia Paparrizou and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/TRICS2010.pdf, TRICS2010.pdf},
year = {2010},
date = {2010-09-06},
urldate = {2010-09-06},
booktitle = {In Proceedings of the TRICS workshop at 6th International Conference on Principles and Practice of Constraint Programming (CP 2010), pages 1–12, St. Andrews, Scotland},
abstract = {The search strategy of a CP solver is determined by the variable and value ordering heuristics it employs and by the branching scheme it follows. Although the effects of variable and value ordering heuristics on search effort have been widely studied, the effects of different branching schemes have received less attention. In this paper we study this effect through an experimental evaluation that includes standard branching schemes such as 2-way, d-way, and dichotomic domain splitting, as well as variations of set branching where branching is performed on sets of values. We also propose and evaluate a generic approach to set branching where the partition of a domain into sets is created using the scores assigned to values by a value ordering heuristic, and a clustering algorithm from machine learning. Experimental results demonstrate that although exponential differences between branching schemes, as predicted in theory between 2-way and d-way branching, are not very common, still the choice of branching scheme can make quite a difference on certain classes of problems. Set branching methods are very competitive with 2-way branching and outperform it on some problem classes. A statistical analysis of the results reveals that our generic clustering-based set branching method is the best among the methods compared.},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
@conference{Balafoutis2010c,
title = { Improving the performance of maxRPC},
author = {Thanasis Balafoutis and Anastasia Paparrizou and Kostas Stergiou and Toby Walsh},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/cp2010.pdf, CP2010.pdf},
year = {2010},
date = {2010-09-06},
urldate = {2010-09-06},
booktitle = {In Proceedings of the 16th International Conference on Principles and Practice of Constraint Programming (CP 2010), LNCS, Vo 6308, pp. 69-83, St Andrews, Scotland},
volume = {6308},
pages = {69-83},
abstract = {Max Restricted Path Consistency (maxRPC) is a local consistency for binary constraints that can achieve considerably stronger pruning than arc consistency. However, existing maxRPC algorithms suffer from overheads and redundancies as they can repeatedly perform many constraint checks without triggering any value deletions. In this paper we propose techniques that can boost the performance of maxRPC algorithms. These include the combined use of two data structures to avoid many redundant constraint checks, and heuristics for the efficient ordering and execution of certain operations. Based on these, we propose two closely related maxRPC algorithms. The first one has optimal O(end 3) time complexity, displays good performance when used stand-alone, but is expensive to apply during search. The second one has O(en 2 d 4) time complexity, but a restricted version with O(end4) complexity can be very efficient when used during search. Both algorithms have O(ed) space complexity when used stand-alone. However, the first algorithm has O(end) space complexity when used during search, while the second retains the O(ed) complexity. Experimental results demonstrate that the resulting methods constantly outperform previous algorithms for maxRPC, often by large margins, and constitute a more than viable alternative to arc consistency.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@article{nokey,
title = {Evaluating and Improving Modern Variable and Revision Ordering Strategies in CSPs},
author = {Thanasis Balafoutis and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/FI.pdf, FI2010.pdf},
year = {2010},
date = {2010-09-01},
urldate = {2010-09-01},
journal = {Fundamenta Informaticae, IOS Press},
volume = {102},
issue = {3-4},
pages = {229-261},
publisher = {IOS Press},
abstract = {A key factor that can dramatically reduce the search space during constraint solving is the criterion under which the variable to be instantiated next is selected. For this purpose numerous heuristics have been proposed. Some of the best of such heuristics exploit information about failures gathered throughout search and recorded in the form of constraint weights, while others measure the importance of variable assignments in reducing the search space. In this work we experimentally evaluate the most recent and powerful variable ordering heuristics, and new variants of them, over a wide range of benchmarks. Results demonstrate that heuristics based on failures are in general more efficient. Based on this, we then derive new revision ordering heuristics that exploit recorded failures to efficiently order the propagation list when arc consistency is maintained during search. Interestingly, in addition to reducing the number of constraint checks and list operations, these heuristics are also able to cut down the size of the explored search tree.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
@conference{Balafoutis2010d,
title = {Adaptive branching for constraint satisfaction problems},
author = {Thanasis Balafoutis and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/ECAI2010.pdf, ECAI2010.pdf},
year = {2010},
date = {2010-08-16},
urldate = {2010-08-16},
booktitle = {In Proceedings of the 19th European Conference on Artificial Intelligence (ECAI 2010), pages 855–860, Lisbon, Portugal},
pages = {855-860},
abstract = {The two standard branching schemes for CSPs are d-way and 2-way branching. Although it has been shown that in theory the latter can be exponentially more effective than the former, there is a lack of empirical evidence showing such differences. To investigate this, we initially make an experimental comparison of the two branching schemes over a wide range of benchmarks. Experimental results verify the theoretical gap between d-way and 2-way branching as we move from a simple variable ordering heuristic like dom to more sophisticated ones like dom/ddeg. However, perhaps surprisingly, experiments also show that when state-of-the-art variable ordering heuristics like dom/wdeg are used then d-way can be clearly more efficient than 2-way branching in many cases. Motivated by this observation, we develop two generic heuristics that can be applied at certain points during search to decide whether 2-way branching or a restricted version of 2-way branching, which is close to d-way branching, will be followed. The application of these heuristics results in an adaptive branching scheme. Experiments with instantiations of the two generic heuristics confirm that search with adaptive branching outperforms search with a fixed branching scheme on a wide range of problems.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@conference{Balafoutis2010,
title = {Conflict directed variable selection strategies for constraint satisfaction problems},
author = {Thanasis Balafoutis and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/SETN2010.pdf, SETN2010.pdf},
year = {2010},
date = {2010-05-04},
urldate = {2010-05-04},
booktitle = {In Proceedings of the 6th Hellenic Conference on Artificial Intelligence (SETN), pages 29–38, Athens},
pages = {29-38},
abstract = {t is well known that the order in which variables are instantiated by a backtracking search algorithm can make an enormous difference to the search effort in solving CSPs. Among the plethora of heuristics that have been proposed in the literature to efficiently order variables during search, a significant recently proposed class uses the learning-from-failure approach. Prime examples of such heuristics are the wdeg and dom/wdeg heuristics of Boussemart et al. which store and exploit information about failures in the form of constraint weights. The efficiency of all the proposed conflict-directed heuristics is due to their ability to learn though conflicts encountered during search. As a result, they can guide search towards hard parts of the problem and identify contentious constraints. Such heuristics are now considered as the most efficient general purpose variable ordering heuristic for CSPs. In this paper we show how information about constraint weights can be used in order to create several new variants of the wdeg and dom/wdeg heuristics. The proposed conflict-driven variable ordering heuristics have been tested over a wide range of benchmarks. Experimental results show that they are quite competitive compared to existing ones and in some cases they can increase efficiency.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@conference{Balafoutis2010e,
title = {Boosting Wikis with Mind Maps in Collaborative Learning Environments},
author = {Thanasis Balafoutis and Panagiotis Pavlos, Eleftherios Pavlos},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/bpp2010.pdf, BPP2010.pdf},
year = {2010},
date = {2010-03-01},
urldate = {2010-03-01},
booktitle = {In Proceedings of the 6th National and International HSSS Conference "Systemic Approaches in Social Structures", Mytilene},
abstract = {Educational communities are recently trying to find ways to supplement traditional methods of teaching and learning, by employing didactic scenarios that expose their students to web based course materials. A web-based platform that has been widely used to support collaborative leaning is Wiki, which allows for easy creation and editing of any number of interlinked web pages with a highly collaborative manner. As a result, Wikis have proven to be of particular interest in supplementing and extending classroom scenarios. However, making the students use the classroom Wiki without further guidance has proven to produce poor results. In educational environments, pieces of knowledge must be easily located and well-structured. It has been shown that the ability of interlinking related articles to break down complex topics or articles is not a trivial task. Students are still in the process of learning how to structure their thoughts in more scientific and organised ways. Thus, a modified Wiki system is necessary and it should be based on the idea of well-structured and easily locatable hypermedia documents. In this paper, we propose the complementary usage of Mind Mapping tools to overcome the drawbacks that Wiki systems have and boost Wiki’s usage in educational environments. The main goal of our work is to provide students, a way to make structured relations between topics when using Wikis. Moreover, in order to show the different levels of interaction between Wikis and mind maps, the case study of the “Greek mythology” project scenario is presented.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@article{Tsinakos2009,
title = {A Comparative Survey On Mind Mapping Tools},
author = {Avgoustos Tsinakos and Thanasis Balafoutis},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/2009.pdf, TOJDE2009.pdf},
issn = {1302-6488},
year = {2009},
date = {2009-06-01},
urldate = {2009-06-01},
journal = {Turkish Online Journal of Distance Education-TOJDE},
volume = {10},
number = {3},
pages = {55-67},
abstract = {Mind Mapping is an important technique that improves the way you takes notes, and enhances your creative problem solving. By using Mind Maps, you can quickly identify and understand the structure of a subject and the way that pieces of information fit together, as well as recording the raw facts contained in normal notes. It can also be used as complementary tools for knowledge construction and sharing. Their suitability as a pedagogical tool for education, e-learning and training, increases their importance. Also, in a world of information overload and businesses struggling to keep up with the place of change, knowledge workers need effective tools to organize, analyze, brainstorm and collaborate on ideas. In resent years, a wide variety of mind mapping software tools have been developed. An often question that comes up, due to this plethora of software tools, is “which is the best mind mapping software?” Anyone who gives you an immediate answer either knows you and your mind mapping activities very well or their answer in not worth a lot. The “best” depends so much on how you use mind maps. In this paper we are trying to investigate different user profiles and to identify various axes for comparison among mind mapping tools that are suitable for a specific user profile, describe each axis and then analyze each tool.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
@workshop{Balafoutis2008,
title = {Experimental Evaluation of Modern Variable Selection Strategies in Constraint Satisfaction Problems},
author = {Thanasis Balafoutis and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/RCRA2008.pdf, RCRA2008.pdf},
year = {2008},
date = {2008-12-13},
urldate = {2008-12-13},
booktitle = {In Proceedings of the 15th RCRA workshop on Experimental Evaluation of Algorithms for Solving Problems with Combinatorial Explosion},
address = {Udine, Italy},
abstract = {Constraint programming is a powerful technique for solving combinatorial search problems that draws on a wide range of methods from artificial intelligence and computer science. Constraint solvers search the solution space either systematically, as with backtracking or branch and bound algorithms, or use forms of local search which may be incomplete. Systematic methods typically interleave search and inference. A key factor that can dramatically reduce the search space is the criterion under which we decide which variable will be the next to be instantiated. Numerous heuristics have been proposed for this purpose in the literature. Recent years have seen the emergence of new and powerful methods for choosing variables during CSP search. Some of these methods exploit information about failures gathered throughout search and recorded in the form of constraint weights, while others measure the importance/impact of variable assignments for reducing the search space. In this paper we experimentally evaluate the most recent and powerful variable ordering heuristics, and new variants of them, over a wide range of academic, random and real world problems. Results demonstrate that heuristics based on failures are in general faster. To be precise, heuristic dom/wdeg and its variants are the dominant heuristics in most instances tried.},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
@workshop{Balafoutis2008c,
title = {Exploiting constraint weights for revision ordering in Arc Consistency Algorithms},
author = {Thanasis Balafoutis and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/ECAIworkshop08.pdf, ECAI2008.pdf},
year = {2008},
date = {2008-06-27},
urldate = {2008-06-27},
booktitle = {In Proceedings of the European Conference on Artificial Intelligence Workshop on Modeling and Solving Problems with Constraints, Patras, Greece},
abstract = {Coarse grained arc consistency algorithms, like AC-3, operate by maintaining a list of arcs (or variables) that records the revisions that are still to be performed. It is well known that the performance of such algorithms is affected by the order in which revisions are carried out. As a result, several heuristics for ordering the elements of the revision list have been proposed. These heuristics exploit information about the original and the current state of the problem, such as domain sizes, variable degrees, and allowed combinations of values, to reduce the number of constraint checks and list operations aiming at speeding up arc consistency computation. Recently, Boussemart et al. proposed novel variable ordering heuristics that exploit information about failures gathered throughout search and recorded in the form of constraint weights. Such heuristics are now considered as the most efficient general purpose variable ordering heuristic for CSPs. In this paper we show how information about constraint weights can be exploited to efficiently order the revision list when AC is applied during search. We propose a number of simple revision ordering heuristics based on constraint weights for arc, variable, and constraint oriented implementations of coarse grained arc consistency algorithms, and compare them to the most efficient existing revision ordering heuristic. Importantly, the new heuristics can not only reduce the numbers of constraints checks and list operations, but also cut down the size of the explored search tree. Results from various structured and random problems demonstrate that some of the proposed heuristics can offer significant speed-ups.},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
@workshop{Balafoutis2008b,
title = {On Conflict-driven variable ordering heuristics},
author = {Thanasis Balafoutis and Kostas Stergiou },
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/csclp2008.pdf, CSCLP2008.pdf},
year = {2008},
date = {2008-06-18},
urldate = {2008-06-18},
booktitle = {In Proceedings of 13th Annual ERCIM International Workshop on Constraint Solving and Constraint Logic Programming, CSCLP-08, Rome, Italy},
abstract = {It is well known that the order in which variables are instantiated by a backtracking search algorithm can make an enormous difference to the search effort in solving CSPs. Among the plethora of heuristics that have been proposed in the literature to efficiently order variables during search, a significant recently proposed class uses the learning-from-failure approach. Prime examples of such heuristics are the wdeg and dom/wdeg heuristics of Boussemart et al. which store and exploit information about failures in the form of constraint weights. The efficiency of all the proposed conflict-directed heuristics is due to their ability to learn though conflicts encountered during search. As a result, they can guide search towards hard parts of the problem and identify contentious constraints. Such heuristics are now considered as the most efficient general purpose variable ordering heuristic for CSPs. In this paper we show how information about constraint weights can be used in order to create several new variants of the wdeg and dom/wdeg heuristics. The proposed conflict-driven variable ordering heuristics have been tested over a wide range of benchmarks. Experimental results show that they are quite competitive compared to existing ones and in some cases they can increase efficiency.},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
@conference{Balafoutis2006,
title = {Algorithms for stochastic CSPs},
author = {Thanasis Balafoutis and Kostas Stergiou},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/10/cp2006scsp.pdf, CP2006.pdf},
year = {2006},
date = {2006-09-25},
urldate = {2006-09-25},
booktitle = {In Proceedings of the 12th International Conference on Principles and Practices of Constraint Programming, LNCS 4204, pages 44–58, Nantes, France},
pages = {44-58},
abstract = {The Stochastic CSP (SCSP) is a framework recently introduced by Walsh to capture combinatorial decision problems that involve uncertainty and probabilities. The SCSP extends the classical CSP by including both decision variables, that an agent can set, and stochastic variables that follow a probability distribution and can model uncertain events beyond the agent’s control. So far, two approaches to solving SCSPs have been proposed; backtracking-based procedures that extend standard methods from CSPs, and scenario-based methods that solve SCSPs by reducing them to a sequence of CSPs. In this paper we further investigate the former approach. We first identify and correct a flaw in the forward checking (FC) procedure proposed by Walsh. We also extend FC to better take advantage of probabilities and thus achieve stronger pruning. Then we define arc consistency for SCSPs and introduce an arc consistency algorithm that can handle constraints of any arity.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
@mastersthesis{Balafoutis2006c,
title = {Algorithms for Stochastic Constraint Satisfaction Problems},
author = {Thanasis Balafoutis},
url = {https://robotics.pme.duth.gr/balafoutis/wp-content/uploads/2022/11/masterThesis.pdf, masterThesis.pdf},
year = {2006},
date = {2006-07-03},
urldate = {2006-07-03},
school = {Artificial Intelligence Laboratory, Department of Information and Communication Systems Engineering, University of the Aegean},
abstract = {In this thesis we studied the Stochastic Constraint Satisfaction Problem, an extension of the Constraint Satisfaction Problem which includes both decision variables (which we can set) and stochastic variables (which follow some probability distribution). The framework is designed to take advantage of the best features of traditional constraint programming, stochastic integer programming, and stochastic satisfiability. It can be used to model a wide variety of decision problems involving uncertainty and probability.
Our work has followed the semantics for stochastic constraint programs based upon policies. These determine how decision variables are set depending on earlier decision and stochastic variables. We have studied the algorithms presented in [Walsh02] and have shown that the FC algorithm suffers from a flaw in the way forward checks are made which can result in erroneous answers. We have corrected the flaw and proposed an improved version of the FC algorithm that achieves stronger pruning.
Moreover, we have described GAC and MAC algorithms for Stochastic CSPs that exploit specialized pruning rules to refute branches of the search tree and thus save search effort. In this way our work generalizes the work of Walsh where only binary constraints were considered.},
keywords = {},
pubstate = {published},
tppubtype = {mastersthesis}
}
Our work has followed the semantics for stochastic constraint programs based upon policies. These determine how decision variables are set depending on earlier decision and stochastic variables. We have studied the algorithms presented in [Walsh02] and have shown that the FC algorithm suffers from a flaw in the way forward checks are made which can result in erroneous answers. We have corrected the flaw and proposed an improved version of the FC algorithm that achieves stronger pruning.
Moreover, we have described GAC and MAC algorithms for Stochastic CSPs that exploit specialized pruning rules to refute branches of the search tree and thus save search effort. In this way our work generalizes the work of Walsh where only binary constraints were considered.
Publications
TraDance: A dataset for assessing the performance of Greek folk dances recognition models Conference In Proceedings of the IEEE/IST International Conference on Imaging Systems & Techniques, Copenhagen, Denmark , 2023. Skeleton-Based Recognition of Traditional Greek Dance Steps using Machine Learning Algorithms Conference In Proceedings of the IEEE/IST International Conference on Imaging Systems & Techniques, Copenhagen, Denmark, 2023. Daedalus Ionospheric Profile Continuation Journal Article In: Geoscientific Instrumentation, Methods and Data Systems, in review, 2022. A Greek QB50 nano-satellite for Upper Atmosphere Studies Miscellaneous Hipparchos. The Hellenic Astronomical Society Newsletter, ISSN: 1790-9252, Volume 2, Issue 12, pages 11-17, 2015, ISSN: 1790-9252. The Importance of Variable Ordering in Constraint Satisfaction Problems Journal Article In: Advanced Computational Technologies, Romanian Academy Publishing House, vol. 322, pp. 206-216, 2012. New Algorithms for max Restricted Path Consistency Journal Article In: Constraints, Springer, vol. 16, no. 4, pp. 372-406, 2011. Adaptive Strategies for Solving Constraint Satisfaction Problems PhD Thesis Artificial Intelligence Laboratory, Department of Information and Communication Systems Engineering, University of the Aegean, 2011. Experimental Evaluation of Branching Schemes for the CSP Workshop In Proceedings of the TRICS workshop at 6th International Conference on Principles and Practice of Constraint Programming (CP 2010), pages 1–12, St. Andrews, Scotland, 2010. Improving the performance of maxRPC Conference In Proceedings of the 16th International Conference on Principles and Practice of Constraint Programming (CP 2010), LNCS, Vo 6308, pp. 69-83, St Andrews, Scotland, vol. 6308, 2010. Evaluating and Improving Modern Variable and Revision Ordering Strategies in CSPs Journal Article In: Fundamenta Informaticae, IOS Press, vol. 102, iss. 3-4, pp. 229-261, 2010. Adaptive branching for constraint satisfaction problems Conference In Proceedings of the 19th European Conference on Artificial Intelligence (ECAI 2010), pages 855–860, Lisbon, Portugal, 2010. Conflict directed variable selection strategies for constraint satisfaction problems Conference In Proceedings of the 6th Hellenic Conference on Artificial Intelligence (SETN), pages 29–38, Athens, 2010. Boosting Wikis with Mind Maps in Collaborative Learning Environments Conference In Proceedings of the 6th National and International HSSS Conference "Systemic Approaches in Social Structures", Mytilene, 2010. A Comparative Survey On Mind Mapping Tools Journal Article In: Turkish Online Journal of Distance Education-TOJDE, vol. 10, no. 3, pp. 55-67, 2009, ISSN: 1302-6488. Experimental Evaluation of Modern Variable Selection Strategies in Constraint Satisfaction Problems Workshop In Proceedings of the 15th RCRA workshop on Experimental Evaluation of Algorithms for Solving Problems with Combinatorial Explosion, Udine, Italy, 2008. Exploiting constraint weights for revision ordering in Arc Consistency Algorithms Workshop In Proceedings of the European Conference on Artificial Intelligence Workshop on Modeling and Solving Problems with Constraints, Patras, Greece, 2008. On Conflict-driven variable ordering heuristics Workshop In Proceedings of 13th Annual ERCIM International Workshop on Constraint Solving and Constraint Logic Programming, CSCLP-08, Rome, Italy, 2008. Algorithms for stochastic CSPs Conference In Proceedings of the 12th International Conference on Principles and Practices of Constraint Programming, LNCS 4204, pages 44–58, Nantes, France, 2006. Algorithms for Stochastic Constraint Satisfaction Problems Masters Thesis Artificial Intelligence Laboratory, Department of Information and Communication Systems Engineering, University of the Aegean, 2006.