Juniper Publishers-Journal of Robotics Engineering

Neural Network Training Using Unscented and Extended Kalman Filter Authored by Denis Pereira de Lima, This work demonstrates the training of a multilayered neural network (MNN) using the Kalman filter variations. Kalman filters estimate the weights of a neural network, considering the weights as a dynamic and upgradable system. The Extended Kalman Filter (EKF) is a tool that has been used by many authors for the training of Neural Networks (NN) over the years. However, this filter has some inherent drawbacks such as instability caused by initial conditions due to linearization and costly calculation of the Jacobian matrices. Therefore, the Unscented Kalman Filter variant has been demonstrated by several authors to be superior in terms of convergence, speed and accuracy when compared to the EKF. Training using this algorithm tends to become more precise compared to EKF, due to its linear transformation technique known as the Unscented Transform (UT). The results pre

Swarm Intelligence Algorithms as the Promising  Tool for Solving Optimisation Problems Authored by Nafrizuan Mat Yahya, A quote by George Bernhard Dantzig, a famous American mathematical scientist who made important contributions to operations research, computer science, economics, and statistics [1]: “True optimisation is the revolutionary contribution of modern research to decision processes”. Optimisation according to the definition of Merriam-Webster Dictionary Merriam-Webster [2] is an act, process, or methodology of making something (as a design, system, or decision) as fully perfect, functional, or effective as possible. In general, optimisation is the process of obtaining either the best minimum or maximum result under specific circumstance Singiresu [3], Xin-She & Suash [4]. Sanghamitra & Sriparna [5], Statnikov et al. [6], Xin-She [7] added that the optimisation process engages with defining and examining objective or fitness function that suits som