In this paper, a new rule based classifier is presented. ADX is an algorithm for inductive learning and for later classification of objects. As is typical for rule systems, an knowledge representation is easy to understand by a human. The power of ADX algorithm is that rules are not too complicated and learning time increases linearly with the size of dataset. The new elements in this work are ...

Rule learning is typically used in solving classification and prediction tasks. However, learning of classification rules can be adapted also to subgroup discovery. Such an adaptation has already been done for the CN2 rule learning algorithm. In previous work this new algorithm, called CN2-SD, has been described in detail and applied to the well known UCI data sets. This paper summarizes the mo...

This paper presents Garss, a new algorithm for rule subset selection based on genetic algorithms, which uses the area under the ROC curve – AUC – as fitness function. Garss is a post-processing method that can be applied to any rule learning algorithm. In this work, Garss is analysed in the context of associative classification, where an association rule algorithm generates a set rules to be us...

evolutionary algorithms are some of the most crucial random approaches tosolve the problems, but sometimes generate low quality solutions. on the otherhand, learning automata are adaptive decision-making devices, operating onunknown random environments, so it seems that if evolutionary and learningautomaton based algorithms are operated simultaneously, the quality of results willincrease sharpl...

Neural network models offer a theoretical testbed for the study of learning at the cellular level. The only experimentally verified learning rule, Hebb’s rule, is extremely limited in its ability to train networks to perform complex tasks. An identified cellular mechanism responsible for Hebbian-type long-term potentiation, the NMDA receptor, is highly versatile. Its function and efficacy are m...

Rule induction systems seek to generate rule sets which are optimal in the complexity of the rule set. This paper develops a formal proof of the NP-Completeness of the problem of generating the simplest rule set (MIN RS) which accurately predicts examples in the training set for a particular type of generalization algorithm algorithm and complexity measure. The proof is then informally extended...

In this paper we will examine the problem of learning an e cient fuzzy logic rule set for the control of the inverted pendulum (with nonlinear dynamics) using an evolutionary algorithm. In particular we compare a two layered rule set with a single fuzzy logic rule set. Furthermore we look at the e ect that di erent choices of objective function (in the evolutionary algorithm) have on the rule s...

A supervised learning rule for Spiking Neural Networks (SNNs) is presented that can cope with neurons that spike multiple times. The rule is developed by extending the existing SpikeProp algorithm which could only be used for one spike per neuron. The problem caused by the discontinuity in the spike process is counteracted with a simple but effective rule, which makes the learning process more ...

The term "learning rule" in neural network theory usually refers to a rule for the plasticity of a given synapse, whereas metaplasticity involves a "metalearning algorithm" describing higher level control mechanisms for apportioning plasticity across a population of synapses. We propose here that the cerebellar cortex may use metaplasticity, and we demonstrate this by introducing the Cerebellar...

In many regression learning algorithms for fuzzy rule bases it is not possible to define the error measure to be optimized freely. A possible alternative is the usage of global optimization algorithms like genetic programming approaches. These approaches, however, are very slow because of the high complexity of the search space. In this paper we present a novel approach where we first create a ...