{"title":"Relational algebra for multi-ranked similarity-based databases","authors":"R. Belohlávek, Vilém Vychodil","doi":"10.1109/FOCI.2013.6602448","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602448","url":null,"abstract":"We present multi-ranked relational model of data which extends the classic Codd's model by considering similarity relations on domains and ranks assigned to values of tuples. The ranks represent degrees to which values in tuples match similarity-based queries. Unlike various single-ranked similarity-based database models where ranks are assigned to whole tuples, in the present model the ranks are assigned to tuple values. As a consequence, the multi-ranked model allows users to directly observe how values in tuples contribute to results of similarity-based queries. We present foundations of the model, relational operations and relational algebra as the primary query language, and its relationship to single-ranked models which have been used in the past. We argue that the multi-ranked model is more suitable for applications in which data analysts require a finer view on results of queries than in the single-ranked model.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116099124","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Influence of selective pressure on quality of solutions and speed of evolutionary mastermind","authors":"J. J. M. Guervós, A. García, C. Cotta, Nuria Rico","doi":"10.1109/FOCI.2013.6602464","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602464","url":null,"abstract":"Mastermind is a puzzle in which a hidden code of length ℓ and made with κ colors has to be discovered via making guesses of the code and receiving hints that express the distance from the guess to the code, in terms of number of symbols in the right position and with the right color. Solutions to these problem are mainly heuristic and thus finding the correct parameters for these solutions has to be done via systematic experimentation. Since diversity in the population is one of the main factors affecting performance, in this paper we will experiment with selective pressure via two different parameters: population size and size of tournament in tournament selection. We will study the influence of them in three different measures: algorithm performance (measured in average number of guesses needed), number of evaluations and time needed to find the solution. We will prove that while, in general, increasing population size improves performance, there is an optimal size over which no further improvement is achieved. On the other hand, tournament size does not have a clear influence on performance, although it influences time needed to find the solution. We will also show that the number of evaluations is correlated positively with time, and it increases with population size so that a trade-off has to be found among solution quality and population size. After evaluating the result of the experiments, we will try to advance a rule of thumb for sizing population for the general MasterMind problem.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132268822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The impact of varying resources available to Iterated Prisoner's Dilemma agents","authors":"D. Ashlock, Eun-Youn Kim","doi":"10.1109/FOCI.2013.6602456","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602456","url":null,"abstract":"The Iterated Prisoner's Dilemma is a simultaneous two-player game widely used in studies on cooperation and conflict. Past work has shown that the choice of representation of evolving agents has a dominant impact on their behavior. In this study we also examine the impact of varying available resources. A variety of different resources can be varied, including amount and type of past information available to agents, number of neurons in a neural net, number of probability levels available to encode a probabilistic strategy, and number of states in a finite state machine. All these resources are shown to have an impact on the character of evolved agents assessed using both play profiles and a total score measure. Play profiles bin the ranges of score space while total score is a global assessment of the type of play that occurs over the course of evolution. The largest effect is found for the probabilistic agents, followed by finite state agents, lookup tables, and finally neural nets exhibit the least effect.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117238058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A review of computational intelligence in RTS games","authors":"R. Lara-Cabrera, C. Cotta, Antonio J. Fernández","doi":"10.1109/FOCI.2013.6602463","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602463","url":null,"abstract":"Real-time strategy games offer a wide variety of fundamental AI research challenges. Most of these challenges have applications outside the game domain. This paper provides a review on computational intelligence in real-time strategy games (RTS). It starts with challenges in real-time strategy games, then it reviews different tasks to overcome this challenges. Later, it describes the techniques used to solve this challenges and it makes a relationship between techniques and tasks. Finally, it presents a set of different frameworks used as test-beds for the techniques employed. This paper is intended to be a starting point for future researchers on this topic.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132175177","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Radial basis functions for multidimensional learning with an application to nondestructive sizing of defects","authors":"S. S. Ahmed, B. Rao, T. Jayakumar","doi":"10.1109/FOCI.2013.6602453","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602453","url":null,"abstract":"A computational intelligence problem with mapping of multiple classes for a given input feature is addressed in this paper. The objective is to classify a vector of class for a given vector of input features. Each class is a member of disjoint set called dimension and hence, it is called multidimensional learning. Dependency between the classes and dimensions are usually not taken into account while constructing independent classifiers for each component class of vector. In this paper, two methods of adaption of radial basis functions (RBF) neural network for multidimensional learning are proposed. In first method, the prototype vector of hidden layer is formed by cluster analysis on instance belong to each class of each dimension. By this way the dependencies of classes is considered. In second method, the prototype vector of hidden layer are formed by cluster analysis on instance belong to each new classes by taking the Cartesian product of each dimension. With this method, the dependency between each dimension is concentrated. A comparison study with these two methods of adaptations with independent uni-dimensional RBF is presented. Studies are carried out with real world multidimensional dataset (with >2 classes in each dimension) obtained from simulated eddy current non-destructive evaluation (NDE) of a stainless steel plate having sub-surface defects of different dimensions. This dataset is used for estimating three characteristics (three dimensions) of defects namely, length, depth and height. The performance evaluation using metrics such as mean accuracy and global accuracy clearly reveals that the proposed multidimensional RBF is superior to the uni-dimensional RBF used individually for each dimensions.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133549116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Evaluation scheduling in noisy environments","authors":"James R. Glenn","doi":"10.1109/FOCI.2013.6602457","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602457","url":null,"abstract":"This paper investigates the problem of scheduling a fixed number of evaluations for genetic algorithms in noisy environments. With a fixed number of evaluations there is a tradeoff between the time the population is allowed to evolve (that is, the number of generations), the size of the population, and the number of samples scheduled per individual in an effort to reduce the effects of noise. This paper focuses mostly on the balance between allocating evaluations to the evolutionary phase versus allocating evaluations to selecting the individual with the highest fitness from the final population (the “champion selection” phase). Several different algorithms for scheduling evaluations during the champion selection phase are compared using a common test function to see how often they find the optimal value. The best algorithm is enhanced to improve its running time. We find the optimal split between the evolutionary and champion selection phases for the selected test function and we examine the effect of varying other parameters such as number of generations (and hence population) and evaluations per individual.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128947708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vaenthan Thiruvarudchelvan, J. Crane, T. Bossomaier
{"title":"Analysis of SpikeProp convergence with alternative spike response functions","authors":"Vaenthan Thiruvarudchelvan, J. Crane, T. Bossomaier","doi":"10.1109/FOCI.2013.6602461","DOIUrl":"https://doi.org/10.1109/FOCI.2013.6602461","url":null,"abstract":"SpikeProp is a supervised learning algorithm for spiking neural networks analogous to backpropagation. Like backpropagation, it may fail to converge for particular networks, parameters and datasets. However there are several behaviours and additional failure modes unique to SpikeProp which have not been explicitly outlined in the literature. These factors hinder the adoption of SpikeProp for general machine learning use. In this paper we examine the mathematics of SpikeProp in detail and identify the various causes of failure therein. The analysis implies that applying certain constraints on parameters like initial weights can improve the rates of convergence. It also suggests that alternative spike response functions could improve the learning rate and reduce the number of convergence failures. We tested two alternative functions and found these predictions to be true.","PeriodicalId":237129,"journal":{"name":"2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126828260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}