The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
Displaying 101 –
120 of
231
Theoretical and experimental studies have shown that traditional training algorithms for Dynamical Recurrent Neural Networks may suffer of local optima solutions, due to the error propagation across the recurrence. In the last years, many researchers have put forward different approaches to solve this problem, most of them being based on heuristic procedures. In this paper, the training capabilities of evolutionary techniques are studied, for Dynamical Recurrent Neural Networks. The performance...
While using automated learning methods, the lack of accuracy and poor knowledge generalization are both typical problems for a rule-based system obtained on a given data set. This paper introduces a new method capable of generating an accurate rule-based fuzzy inference system with parameterized consequences using an automated, off-line learning process based on multi-phase evolutionary computing and a training data covering algorithm. The presented method consists of the following steps: obtaining...
The paper presents a new Artificial Neural Network (ANN) encoding method called Assembler Encoding (AE). It assumes that the ANN is encoded in the form of a program (Assembler Encoding Program, AEP) of a linear organization and of a structure similar to the structure of a simple assembler program. The task of the AEP is to create a Connectivity Matrix (CM) which can be transformed into the ANN of any architecture. To create AEPs, and in consequence ANNs, genetic algorithms (GAs) are used. In addition...
This paper addresses the task of learning classifiers from streams of labelled data. In this case we can face the problem that the underlying concepts can change over time. The paper studies two mechanisms developed for dealing with changing concepts. Both are based on the time window idea. The first one forgets gradually, by assigning to the examples weight that gradually decreases over time. The second one uses a statistical test to detect changes in concept and then optimizes the size of the...
The extraction of logical rules from data has been, for nearly fifteen years, a key application of artificial neural networks in data mining. Although Boolean rules have been extracted in the majority of cases, also methods for the extraction of fuzzy logic rules have been studied increasingly often. In the paper, those methods are discussed within a five-dimensional classification scheme for neural-networks based rule extraction, and it is pointed out that all of them share the feature of being...
A new method of parameter estimation for an artificial neural network inference system based on a logical interpretation of fuzzy if-then rules (ANBLIR) is presented. The novelty of the learning algorithm consists in the application of a deterministic annealing method integrated with ε-insensitive learning. In order to decrease the computational burden of the learning procedure, a deterministic annealing method with a "freezing" phase and ε-insensitive learning by solving a system of linear inequalities...
In this paper the possibilities are discussed for training statistical pattern recognizers based on a distance representation of the objects instead of a feature representation. Distances or similarities are used between the unknown objects to be classified with a selected subset of the training objects (the support objects). These distances are combined into linear or nonlinear classifiers. In this approach the feature definition problem is replaced by finding good similarity measures. The proposal...
This paper deals with the possibility of using a feedforward neural network to test the discrepancies between a real astronomical image and a predefined template. This task can be accomplished thanks to the capability of neural networks to solve a nonlinear approximation problem, i.e. to construct an hypersurface that approximates a given set of scattered data couples. Images are encoded associating each of them with some conveniently chosen statistical moments, evaluated along the axes; in this...
Fuzzy data mining by means of the fuzzy decision tree method enables the construction of a set of fuzzy rules. Such a rule set can be associated with a database as a knowledge base that can be used to help answering frequent queries. In this paper, a study is done that enables us to show that classification by means of a fuzzy decision tree is equivalent to the generalized modus ponens. Moreover, it is shown that the decision taken by means of a fuzzy decision tree is more stable when observation...
We have shown a model of fuzzy neural network that is able to infer the relations associated to the transitions of a fuzzy automaton from a fuzzy examples set. Neural network is trained by a backpropagation of error based in a smooth derivative [1]. Once network has been trained the fuzzy relations associated to the transitions of the automaton are found encoded in the weights.
In this paper, an architecture of fuzzy neural networks is proposed to find a real root of a dual fuzzy polynomial (if exists) by introducing a learning algorithm. We proposed a learning algorithm from the cost function for adjusting of crisp weights. According to fuzzy arithmetic, dual fuzzy polynomials can not be replaced by a fuzzy polynomials, directly. Finally, we illustrate our approach by numerical examples.
This article presents an XML[2] based language for the specification of objects in the Soft Computing area. The design promotes reuse and takes a compositional approach in which more complex constructs are built from simpler ones; it is also independent of implementation details as the definition of the language only states the expected behaviour of every possible implementation. Here the basic structures for the specification of concepts in the Fuzzy Logic area are described and a simple construct...
Currently displaying 101 –
120 of
231