The paper presents a two-level learning method for the design of the Beta Basis Function Neural Network BBFNN. A Genetic Algorithm is employed at the upper level to construct BBFNN, while the key learning parameters :the width, the centers and the Beta form are optimised using the gradient algorithm at the lower level. In order to demonstrate the effectiveness of this hierarchical learning algo...

This paper shows that Error-Driven Constraint Demotion (EDCD), an errordriven learning algorithm proposed by Tesar (1995) for Prince and Smolensky’s (1993) version of Optimality Theory, can fail to converge to a totally ranked hierarchy of constraints, unlike the earlier non-error-driven learning algorithms proposed by Tesar and Smolensky (1993). The cause of the problem is found in Tesar’s use...

One of the most important and valuable goal of software development life cycle is software cost estimation or SCE. During the recent years, SCE has attracted the attention of researchers due to huge amount of software project requests. There have been proposed so many models using heuristic and meta-heuristic algorithms to do machine learning process for SCE. COCOMO81 is one of the most popular...

In this paper, a novel learning algorithm termed Hybrid Online Sequential Extreme Learning Machine (HOSELM) is proposed. The proposed HOS-ELM algorithm is a fusion of the Online Sequential Extreme Learning Machine (OS-ELM) and the Minimal Resource Allocation Network (MRAN). It is capable of reducing the number of hidden nodes in Single-hidden Layer Feed-forward Neural Networks (SLFNs) with Radi...

This article shows that Error-Driven Constraint Demotion (EDCD), an error-driven learning algorithm proposed by Tesar (1995) for Prince and Smolensky’s (1993/2004) version of Optimality Theory, can fail to converge to a correct totally ranked hierarchy of constraints, unlike the earlier non-error-driven learning algorithms proposed by Tesar and Smolensky (1993). The cause of the problem is foun...

the hybrid fuzzy differential equations have a wide range of applications in science and engineering. we consider the problem of nding their numerical solutions by using a novel hybrid method based on fuzzy neural network. here neural network is considered as a part of large eld called neural computing or soft computing. the proposed algorithm is illustrated by numerical examples and the resu...

In a potentially congested network, random early detection (RED) active queue management (AQM) proved effective in improving throughput and average queuing delay. The main disadvantage of RED is its sensitive parameters that are impossible to estimate perfectly and adjust manually because of the dynamic nature of the network. For this reason, RED performs differently during different phases of ...

Over the years complexity theorists have proposed many structural parameters to explain the surprising efficiency of conflictdriven clause-learning (CDCL) SAT solvers on a wide variety of large industrial Boolean instances. While some of these parameters have been studied empirically, until now there has not been a unified comparative study of their explanatory power on a comprehensive benchmar...

The results of this dissertation are two-fold. On the one hand, inductive learning techniques are extended and two new inference algorithms for inferring nondeterministic, and universal, respectively, finite-state automata are presented. On the other hand, certain learning techniques are employed and enhanced to semi-automatically infer communicating automata (also called design models in the s...

In this paper, we leverage both deep learning and conditional random fields (CRFs) for sequential labeling. More specifically, we propose a mixture objective function to predict labels either independent or correlated in the sequential patterns. We learn model parameters in a simple but effective way. In particular, we pretrain the deep structure with greedy layer-wise restricted Boltzmann mach...