Lee, Sin Wee and Palmer-Brown, Dominic (2006) ‘Modal Learning in a Neural Network’, Proceedings of the AC&T, pp. 42-47.
Use this permanent URL when citing or linking to this resource in ROAR.
This paper presents an application of the snap-drift modal learning algorithm developed in recent years by Lee and Palmer-Brown (Lee, 2004a). The application involves phrase recognition using a set of phrases from the Lancaster Parsed Corpus (LPC) (Garside, 1987). The learning algorithm is the classifier version of snap-drift. The twin modes of minimalist learning (snap) and slow drift towards the input pattern are applied alternately. Each neuron of the Snap-Drift Neural Network (SDNN) swaps between snap and drift modes when declining performance is indicated on that particular node, so that each node has its learning mode toggled independently of the other nodes. Learning on each node is also reinforced by enabling learning with a probability that decreases with increasing performance. The simulations demonstrate that learning is stable, and the results have consistently shown similar classification performance and advantages in terms of speed in comparison with a Multilayer Perceptron (MLP) and back-propagation neural networks applied to the same problem.
|Divisions:||Schools > Architecture Computing and Engineering, School of|
|Additional Information:||Citation: Lee, S.W; Palmer-Brown, D. (2006) ‘Modal Learning in a Neural Network’ Proceedings of the AC&T, pp.42-47..|
|Date Deposited:||09 Jun 2010 12:25|
|Item Type:||Conference or Event Item (Paper)|
|Creators:||Lee, Sin Wee and Palmer-Brown, Dominic|
|Last Modified:||27 Sep 2012 11:58|
|Depositing User:||Stephen Grace|