In the ANN , they suppose there are sufficient knowledge about the correct results of the problem?
I work to solve a problem without any information about the sources, by using one of the Artificial Neural Networks.
There is neural network work without any prior information about the targets?
Hello,
there are several learning algorithms for artificial neural networks that do not require labeled answers for the training data. This type of problem is known as unsupervised learning. Some of the more prominent unsupervised network architectures are autoencoders and deep belief networks. If you want to do something like PCA / ICA or blind source separation, it might also be worthwhile to look at the Oja learning rule and its variants.
Best regards,
Christian
Hello,
there are several learning algorithms for artificial neural networks that do not require labeled answers for the training data. This type of problem is known as unsupervised learning. Some of the more prominent unsupervised network architectures are autoencoders and deep belief networks. If you want to do something like PCA / ICA or blind source separation, it might also be worthwhile to look at the Oja learning rule and its variants.
Best regards,
Christian
Hi Hussein,
Yes, there is such type of NN that can perform a clustering task. The best example algorithm is "Self-organizing map" or "SOM". For more information, please have a look at this link:
https://en.wikipedia.org/wiki/Self-organizing_map
HTH.
Samer
See more. For example the two folowing books have good texts about unsupervised learning: "The elements of statistical learning" (T. Hastie, R. Tibshirani, J. Friedman) and "Pattern recognitions and neural networks" (B.D. Ripley).
You need to study supervised learning and unsupervised learning. there are several model available for both learning.
unsupervised feature learning is that if we can get our algorithms to learn from ”unlabeled” data, then we can easily obtain and learn from massive amounts of it.
- Badges
- Science topic