The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The following topics are dealt with: artificial neural network; evolutionary computing; fuzzy logic; applied computational intelligence; and fuzzy set theory.
In this paper, a combination of fuzzy system models and simulated annealing are used to predict Mackey-Glass time series with different levels of added noise by searching for the best configuration of the fuzzy system. Simulated annealing is used to optimise the parameters of the antecedent and the consequent parts of the fuzzy system rules under singleton and non-singleton fuzzifications for both...
The assessment of mammographie risk analysis is an important issue in the medical field. Various approaches have been applied in order to achieve a higher accuracy in such analysis. In this paper, an approach known as Extreme Learning Machines (ELM), is employed to generate a single hidden layer neural network based classifier for estimating mammographic risk. ELM is able to avoid problems such as...
As part of this paper we are highlighting several - in our opinion- important aspects of type-2 fuzzy logic systems which seem important to its future development and application. It is the aim of the paper to provide more questions and more suggestive points than actual answers. With type-2 fuzzy logic and its application to modelling and handling uncertainty still a very young area of research,...
This paper describes the winner algorithm of the Car Setup Optimization Competition that took place in EvoStar (2010). The aim of this competition is to create an optimization algorithm to fine tune the parameters of a car in the The Open Racing Car Simulator (TORCS) video game. There were five participants of the competition plus the two algorithms presented by the organizers (that do not take part...
Word vectors and sets of words are used in a wide range of text-based applications. Yet these word sets are often chosen on an ad hoc basis. In this study, we examine two text-based applications that use word sets and in both cases find that classification performance can be optimised using a fairly simple genetic algorithm. The first study is in authorship attribution, the second one is sentiment...
The vast majority of fuzzy logic systems and applications employ singleton fuzzification because of its simplicity and speed of computation which allows for real time operation. However, using singleton fuzzification assumes that the input measurements are clean signals with no noise or uncertainty associated with them. The vast majority real world applications have high values of noise and uncertainty...
Ants in conventional ant colony optimization (ACO) algorithms use pheromone to communicate. Usually, this indirect communication leads the algorithm to a stagnation behaviour, where the ants follow the same path from early stages. This occurs because high levels of pheromone are developed, which force the ants to follow the same corresponding trails. As a result, the population gets trapped into a...
Coevolutionary algorithms employ collaboration methods in assessing the fitness of solutions. In this paper, we explore four different collaboration methods for coevolving technical trading rules for entering, and exiting long and short positions, and stop loss rules for long and short positions respectively. Our results show that our problem is sensitive to the collaboration method being used and...
I propose a new characterization of the types of problems for which computational intelligence (CI) tends to be used, namely the identification of approximate abstractions. I then suggest that equity markets provide a challenging example for CI. Because markets are inherently adaptive, they pose a more difficult problem than traditional CI domains. I discuss my experience teaching a CI class that...
Since the rise in popularity of type-2 fuzzy logic systems a reoccurring problem has been the transition from existing type-1 fuzzy sets to type-2 fuzzy sets. As part of this paper, we are providing an overview of the said transition and discuss the requirement for an operation which allows to generate a single type-2 fuzzy set from multiple type-1 fuzzy sets. In this context, we discuss the limitations...
Data-to-text natural language generation techniques do not currently impart deep meaning in their output and leave it to an expert user to draw causal inferences. Frequently, the expert is adding meaning that would be present in data sources that could be made available to the NLG system. As the system is intended to convey as much information as possible, it seems counterintuitive to require the...
The operation of drilling rigs is highly expensive. It is therefore important to be able to identify and analyse variables affecting rig operations. We investigate the use of Genetic Algorithms and Ant Colony Optimisation to induce a Bayesian Network model for the real world problem of Rig Operations Management and confirm the validity of our previous model. We explore the relative performances of...
This paper modifies the replacement and update scheme in MOEA/D-DE developed in for dealing with constraints in multiobjective optimization problems. The modified scheme introduces a penalty function to penalize infeasible solutions. The penalty function uses a threshold to control the amount of penalty to infeasible solutions. Experimental results have shown that this penalty method is very promising.
In recent years, significant research attention has been paid to evolving self-learning checkers players. Fogel's Blondie24 has been very successful in this field and has inspired other researchers to further develop this area. In this paper we address the question of whether piece difference is an important factor in the Blondie24 architecture. Although this issue has been addressed before, this...
Authorship Attribution is the problem of determining the authorship of one or more texts. Applications include disputed authorship, or deciding which of a collection of pieces of text were by the same author. A popular and successful approach is to characterize a specific author in terms of the usage pattern of function words. These are common words that are unrelated to subject matter, and tend to...
Different genetic operators suit different problems. Using several crossover operators should be an effective approach for improving the performance of an evolutionary algorithm. This paper studies the effect of the use of two crossover operators on MOEA/D-DRA for multi-objective optimization. It considers two crossover operators, namely, simplex crossover operator (SPX) and center of mass crossover...
Inspired by the major principles of gene regulation and cellular interactions in multi-cellular development, this paper proposes a distributed self-organizing algorithm for multi-robot shape formation. In this approach, multiple robots are able to self-organize themselves into complex shapes driven by the dynamics of a gene regulatory network model. Particularly, no predefined global coordinate system...
Temporal difference learning (TDL) is perhaps the most widely used reinforcement learning method and gives competitive results on a range of problems, especially when using linear or table-based function approximators. However, it has been shown to give poor results on some continuous control problems and an important question is how it can be applied to such problems more effectively. The crucial...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.