Peter Edwards

AAAI Conference 2011 Conference Paper

Assessing Quality in the Web of Linked Sensor Data

• Chris Baillie
• Peter Edwards
• Edoardo Pignotti

Assessing the quality of sensor data available on the Web is essential in order to identify reliable information for decision-making. This paper discusses how provenance of sensor observations and previous quality ratings can influence quality assessment decisions.

ECAI Conference 2004 Conference Paper

Context-Aware Personalised Service Delivery

• Edoardo Pignotti
• Peter Edwards
• Gunnar Aastrand Grimnes

ECAI Conference 1998 Conference Paper

Implicit Feature Selection with the Value Difference Metric

• Terry R. Payne
• Peter Edwards

NeurIPS Conference 1992 Conference Paper

Synaptic Weight Noise During MLP Learning Enhances Fault-Tolerance, Generalization and Learning Trajectory

• Alan Murray
• Peter Edwards

We analyse the effects of analog noise on the synaptic arithmetic during MultiLayer Perceptron training, by expanding the cost func(cid: 173) tion to include noise-mediated penalty terms. Predictions are made in the light of these calculations which suggest that fault tolerance, generalisation ability and learning trajectory should be improved by such noise-injection. Extensive simulation experiments on two distinct classification problems substantiate the claims. The re(cid: 173) sults appear to be perfectly general for all training schemes where weights are adjusted incrementally, and have wide-ranging implica(cid: 173) tions for all applications, particularly those involving "inaccurate" analog neural VLSI.