what is the differences between classifires and associative memories ?
I think they may use different mathematical apparatus and have distinct purposes.
E.g., the associative memories may not only classify some objects but to produce some "new" data due to internal "incorrect" representations.
As a metaphore, classifiers perform pattern recognition or data clustering by associating a label with a given input pattern, whereas associative memories learn associations between input patterns and expected output ones (some kind of look-up table).
The classifiers are used to classify data into some predefined class labels
such as 'Class A', 'Class B' etc...The class labels are acquired with some domain knowledge, proved techniques or existing methods. Whereas Associative memories
identify associations between two or more set of elements acquired from some learning rule or data mining techniques.
As an associative memory-like neural network extensively used to solve robot motion control problems, let us cite the biologically-inspired Cerebellar Model Articulation Controller (CMAC) invented by Albus in 1975. For further details, follow:
- Laufer et al., "Efficient Recursive Least Squares Methods for the CMAC Neural Network", 2011 - http://www.ijmlc.org/papers/04-L0011%260062.pdf
- Laufer et al., "Kernel Recursive Least Squares for the CMAC Neural Network", 2013 - http://ijcte.org/papers/729-L062.pdf
- Badges
- Science topic
- Similar topics
- Computer Science
- Data Mining