T. Kohonen. Self-Organisation and Associative Memory. Springer, 3rd edition, 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and decoder is often designated as the training set, and the rst M input vectors of the training data set are normally used to initialize all the neurones. With this general structure, various learning algorithms have been designed and developed such as Kohonen s self organizing feature mapping [10,13,18,33,52,70], competitive learning [1,54,55,65] frequency sensitive competitive learning [1,10] fuzzy competitive learning [11,31,32] general learning [25,49] and distortion equalized fuzzy competitive learning [7] and PVQ (predictive VQ) neural networks [46] Let W (t) be the weight vector of the ith ....

...., 2.24) where D ## ### #x = # ##. Hence, the learning rule can be designed as follows: 2.25) for the winning neurone i, and #x = # (2.26) for the other (M 1) neurones. This algorithm can also be classi ed as a variation of Kohonen s selforganizing neural network [33]. Around the competitive learning scheme, fuzzy membership functions are introduced to control the transition from soft to crisp decisions during the code book design process [25,31] The essential idea is that one input vector is assigned to a cluster only to a certain extent rather than either ....

T. Kohonen, Self-Organisation and Associative Memory, Springer, Berlin, 1984.

....of the representation and to extract classifiable information from the properties of the representation. The use of the K means algorithm is perhaps a poor choice due to the inconsistent nature of its results. Future work will investigate more sophisticated approaches such as self organising maps [6]. The ability to generalise to novel views within these representations is also dependent upon the characteristics of such representations the chosen discrete point model possibly represents the simplest approach and the 2 There were 17 with dark hair, 8 wearing glasses, 4 with beards, 1 with ....

T. Kohonen. Self-Organisation and Associative Memory. Springer-Verlag, Heigelberg, 1984.

....1998) By contrast, the model outlined here does not require any assumptions concerning the distribution of patterns, and it discriminates whether a particular pattern was presented previously rather than whether the pattern is typical. The proposed network differs also from the novelty filter (Kohonen, 1989), which determines which bits of the delivered pattern differ from the closest familiar pattern. The proposed model has just a single output but may discriminate familiarity for many more patterns than a novelty filter. Although the information processing in the proposed network is similar to that ....

....pattern differ from the closest familiar pattern. The proposed model has just a single output but may discriminate familiarity for many more patterns than a novelty filter. Although the information processing in the proposed network is similar to that in a novelty detector (Kohonen et al. 1974; Kohonen, 1989), the novelty detector is an abstract model of a single neuron, with correspondingly limited storage capacity. The proposed model is a network of neurons with a very large storage capacity. Furthermore, the performance of the novelty detector was analyzed for an abstract case in which each ....

[Article contains additional citation context not shown here]

Kohonen T (1989) Self-organisation and Associative Memory. (3rd ed.). Springer-Verlag, Heidelberg.

....gives a robust, noise tolerant method of landmark detection. 2.1.1. The RCE network The role of the classifier is to classify sensory perceptions such that noise and small perturbations do not affect the robot s ability to recognise an already mapped location. Self organising feature maps (e.g. [6]) are a possibility and have successfully been used for this purpose [12] However, one particular aspect of the work presented here was to scale up a navigation mechanism from laboratory sized environments to potentially unbounded environments. A network of finite size is problematic in such ....

T. Kohonen, Self-Organisation and Associative Memory, Springer, Berlin, 1988.

....the stochastic recursion to converge to the stable points of the deterministic dynamics are satisfied. Like the linear PCA neuron, but unlike other unsupervised structure finding networks such as competitive nets [Rumelhart Zipser 1985] learning vector quantisers or topographic feature maps [Kohonen 1989], individual ISO neurons self organise independently: without (lateral) interaction with one another. Consequently, a set of ISO neurons adapting to the same pattern vectors will have no ordering or topographic neighbourhood relationships. ISO neurons in multiple level stacks are oblivious to this ....

Self-organisation and Associative Memory (third edition), Springer.

....algorithm is to guide learning in mfns in such a way that the hidden unit activation patterns produced by instances of the same class become more similar to each other. To achieve this aim, the traditional mfn architecture (Rumelhart et al. 1986) is combined with self organising maps (soms) [Kohonen, 1989]: each hidden layer of the mfn is associated with one som (see Figure 1) During training of the weights in the mfn, the corresponding som is trained on the hidden unit activation patterns. In addition to error back propagation (Rumelhart et al. 1986) information from the soms is included when ....

.... as learning vectors for the corresponding som (a hiddenunit activation learning vector is denoted by V hidden ) After a number of training cycles (i.e. presentations of all training instances to the mfn) each som, to a certain extent, develops self organisation (since each som is trained with Kohonen s (1989) som learning algorithm) The result of the self organisation is to be translated into classification information, i.e. each som element must be provided with a class label (one of the output classes of the task) The procedure for determining the class labels for a som is as follows. After each ....

Kohonen, T. (1989). Self-organisation and Associative Memory. Berlin: Springer Verlag.

....to the robot s actions. Why not using another ANN architecture for the experiments Since part of the learning task in our experiments was to learn the topography of the environment relative to landmarks, one might wonder why we did not use a self organising map such as, e.g. Kohonen nets [23] or other models developed previously for robotic tasks (e.g. 30] 46] The reason is simply that the associations we want to make in the sensor actuator state are point like, that is they have no topographical relationships in the sensor actuator vector space (this is demonstrated in figure 15 ....

Kohonen T. (1989), `Self-organisation and associative memory', Springer-Verlag, Berlin, 3rd Edition.

No context found.

T. Kohonen. Self-Organisation and Associative Memory. Springer, 3rd edition, 1990.

No context found.

Teuvo Kohonen. Self-Organisation and Associative Memory. Springer, 3rd edition, 1990.

No context found.

Kohonen, T. (1984), Self organisation and associative memory, Springer Verlag, New York, USA.

No context found.

Kohonen T. 1989. Self-organisation and associative memory. 3rd ed. Heidelberg: Springer-Verlag.

No context found.

T. Kohonen. Self-Organisation and Associative Memory. Springer-Verlag, Berlin, 3rd edition, 1988.

No context found.

T. Kohonen. Self-Organisation and Associative Memory. Springer-Verlag, 1984.

No context found.

T. Kohonen. Self-Organisation and Associative Memory. Springer-Verlag, 1988. 2nd edition.

No context found.

Kohonen T. Self-Organisation and Associative Memory. Springer-Verlag, Berlin, 1984

No context found.

Kohonen, T. (1984) Self Organisation and Associative Memory. Springer-Verlag, Berlin.

No context found.

T. Kohonen, Self--Organisation and Associative Memory. Springer, 3rd edition, 1990.

No context found.

T. Kohonen, Self-organisation and associative memory, Springer-Verlag, Berlin, 1988.

No context found.

Kohonen, T. (1984) Self Organisation and Associative Memory. Springer-Verlag, Berlin.

No context found.

Kohonen,T. (1989)."Self Organisation and Associative memory". Third Edition Springer Verlag.

No context found.

Kohonen, T. (1984) Self Organisation and Associative Memory. Springer-Verlag, Berlin.

No context found.

T Kohonen. Self-Organisation and Associative Memory. Springer-Verlag, 3rd edition, 1989. 185

No context found.

T. Kohonen, "Self-Organisation and Associative Memory," Springer-Verlag, New York, 1989.

No context found.

T. Kohonen. Self Organisation and Associative Memory. Springer-Verlag, Berlin, 1984.

No context found.

T. Kohonen, "Self Organisation and Associative Memory", Springer Verlag, Berlin 1988.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC