Prediction- and simulation-error based perceptron training: Solution space analysis and a novel combined training scheme

Previous papers have noted the difficulty in obtaining neural models which are stable under simulation when trained using prediction-error-based methods. Here the differences between series-parallel and parallel identification structures for training neural models are investigated. The effect of the error surface shape on training convergence and simulation performance is analysed using a standard algorithm operating in both training modes. A combined series-parallel/parallel training scheme is proposed, aiming to provide a more effective means of obtaining accurate neural simulation models. Simulation examples show the combined scheme is advantageous in circumstances where the solution space is known or suspected to be complex. (c) 2006 Elsevier B.V. All rights reserved.

Improving neural network training solutions using regularisation

This paper describes the application of regularisation to the training of feedforward neural networks, as a means of improving the quality of solutions obtained. The basic principles of regularisation theory are outlined for both linear and nonlinear training and then extended to cover a new hybrid training algorithm for feedforward neural networks recently proposed by the authors. The concept of functional regularisation is also introduced and discussed in relation to MLP and RBF networks. The tendency for the hybrid training algorithm and many linear optimisation strategies to generate large magnitude weight solutions when applied to ill-conditioned neural paradigms is illustrated graphically and reasoned analytically. While such weight solutions do not generally result in poor fits, it is argued that they could be subject to numerical instability and are therefore undesirable. Using an illustrative example it is shown that, as well as being beneficial from a generalisation perspective, regularisation also provides a means for controlling the magnitude of solutions. (C) 2001 Elsevier Science B.V. All rights reserved.

The contribution of visual feedback to visuomotor adaptation: How much and when?

We investigated the role of visual feedback in adapting to novel visuomotor environments. Participants produced isometric elbow torques to move a cursor towards visual targets. Following trials with no rotation, participants adapted to a 60 degrees rotation of the visual feedback before returning to the non-rotated condition. Participants received continuous visual feedback (CF) of cursor position during task execution or post-trial visual feedback (PF). With training, reductions of the angular deviations of the cursor path occurred to a similar extent and at a similar rate for CF and PF groups. However, upon re-exposure to the non-rotated environment only CF participants exhibited post-training aftereffects, manifested as increased angular deviation of the cursor path, with respect to the pre-rotation trials. These aftereffects occurred despite colour cues permitting identification of the change in environment. The results show that concurrent feedback permits automatic recalibration of the visuomotor mapping while post-trial feedback permits performance improvement via a cognitive strategy. (C) 2008 Elsevier B.V. All rights reserved.