> A more natural way seems to me: predict what state the PRS is in the next > time a prediction failure (of the base level) happens. The PRS can be seen > as containing all the information about the future (also future PRSes) > until a prediction failure. Apparantly the environment then switches to > some new unforseen behaviour. Until then the environment did not contain > any new information; it can be compressed to the PRS (and the Abstraction > and Prediction neural networks which are fixed during run time).
A correction: The Prediction neural network computes the predicted next input on basis of the most recent input also (Pr: C, I -> Ipred). Therefor, even if the prediction succeeds, it can be that the PRS does not contain all the information to construct the future input. The sequence untill a prediction failure is compressible to the PSR plus the present input. Because of this the PSR can be viewed/operate as an abstract classification without detail information (which is stored in the present input I) (although the PSR can also hold the detail information in very simple environments) , so it's not an unimportant point. second: The PRS that is predicted is the PRS (= context C) after the prediction failure. By the way: I'm going to use BPTT (backpropagation through time) to train the recurrent neural network. According to Schmidhuber at IDSIA there is a better recurrent neural network system Long Short Term Memory (LSTM). This network + training technique is claimed to be able to look further back (>1000 steps in stead of just 10 for BPTT). Is anyone familiair with this? Can it be put in the same layering architecture that I am aiming at, i.e. does it have an abstraction state? Bye, Arnoud ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
