Efficient incremental construction of RBF networks using quasi-gradient method

Reiner, Philip; Wilamowski, Bogdan M.

doi:10.1016/j.neucom.2014.05.082

Cited by 28 publications

(12 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The measure of performance between the algorithms is provided by the AUC, which indicates how much a model is capable of distinguishing between classes: a model whose predictions are 100% wrong has an AUC of 0.0; one whose predictions are 100% correct has an AUC of 1.0. We compare the performance of the iSVM with two well-known incremental algorithms, the incremental Logistic Regression (iLR) [37] and an incremental artificial neural network, the incremental Radial Basis Function network (iRBF) [38]. The testing network used for the validation is composed by a requester interacting with nodes, as providers, that implement each a different behaviour, from benevolent to all of the seven possible attacks.…”

Section: B Simulation Results For ML Algorithmsmentioning

confidence: 99%

Trust-Related Attacks and Their Detection: A Trust Management Model for the Social IoT

Marche

Nitti

2021

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

The integration of social networking concepts into the Internet of Things (IoT) has led to the so called Social Internet of Things paradigm, according to which the objects are capable of establishing social relationships in an autonomous way with respect to their owners. Within this scenario, "things" interact opportunistically with their peers to seek needed services. However, attacks and malfunctions in the IoT can outweigh any of its benefits if not handled adequately. In this paper, we focus on the possible types of trust attacks that can affect the IoT and propose a trust management model able to overcome all the analyzed attacks. Simulations show how the proposed model can effectively isolate almost any malicious nodes in the network at the expense of an increase in the number of transactions needed for the model to converge.

show abstract

Section: B Simulation Results For ML Algorithmsmentioning

confidence: 99%

Trust-Related Attacks and Their Detection: A Trust Management Model for the Social IoT

Marche

Nitti

2021

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

“…Moreover, one of the recent surveys on a comparison of incremental online machine learning techniques [17], covers a broad range of algorithms. According to their results, we are also comparing the proposed method with Incremental Support Vector Machine (ISVM) [18], [19], [20], incremental decision tree based on C4.5 [21] and ID3, incremental Bayesian classifier [22], Online Random Forest (ORF) [23] and Multi-Layer Neural Networks for classification with localist models like Radial Basis Functions (RBF) which work reliably in incremental settings [24], [25].…”

Section: Methodsmentioning

confidence: 99%

Handling Unforeseen Failures Using Argumentation-Based Learning

Ayoobi

Cao

Verbrugge

et al. 2019

2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)

View full text Add to dashboard Cite

General Purpose Service Robots operate in different environments of a dynamic nature. Even the robot's programmer cannot predict what kind of failure conditions a robot may confront in its lifetime. Therefore, general purpose service robots need to efficiently handle unforeseen failure conditions. This requires the capability of handling unforeseen failures while the robot is performing a task. Existing research typically offers special-purpose solutions depending on what has been foreseen at the design time. In this research, we propose a general purpose argumentation-based architecture which is able to autonomously recover from unforeseen failures. We compare the proposed method with existing incremental online learning methods in the literature. The results show that the proposed argumentation-based learning approach is capable of learning complex scenarios faster with a lower number of observations. Moreover, the final precision of the proposed method is higher than other methods.

show abstract

“…In nonlinear control problem, the radial basis function (RBF) network is usually used as a tool for modeling nonlinear system because of its good capabilities in function approximation. In this paper, the unknown δ ( ) x u is approximated by the RBF network (Liu & Zhang, 2013;Reiner & Wilamowski, 2015;Yucelen & Calise, 2011) In fact, an active control system may inevitably suffer from time delay problem. Many factors including measurement of system variables, controller calculation, processes for actuators to build up the required control force, etc., may result in non-synchronization of control force.…”

Section: Finite-time H 1 Adaptive Fault-tolerant Control Designmentioning

confidence: 99%