Well, in principle, Novamente is intended to be able to learn from zippo -- i.e. NO explicitly encoded knowledge.
However, the architecture does support the loading-in of prefab knowledge. Whether, and in what ways, it is possible to introduce prefab knowledge into a learning AGI system, without screwing up the learning and self-formation process, is something we'll have to find out through experience. Right now we are using a simplified Novamente for bioinformatic data analysis (the Biomind AI Engine), and for this purpose we've loaded in a dozen public biology databases (a painful process of crosslinking them then exporting them as XML formatted in Novamente's DTD) We have considered loading some linguistic information into Novamente later on -- e.g. WordNet, XTag. A big question is whether this can really help with language learning or not. My intuition is that it may be useful to load this data into the system AFTER it has already learned some rudiments of language on its own, to accelerate the transition from linguistic rudiments to linguistic master. But at our present sluggish pace we are 1-2 years from being ready to teach the beast language ;-( [This could be accelerated to probably 9 months if more funding appeared!] Also, the Mizar corpus of mathematical knowledge (www.mizar.org) would be a good candidate -- kind of a brain lobe full of mathematical proofs! In short, in Novamente, loading in of prefab knowledge is an optional extra to be experimented with, not a key part of the design or the teaching process. -- Ben G > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of Brad Wyble > Sent: Tuesday, February 25, 2003 3:20 PM > To: [EMAIL PROTECTED] > Subject: [agi] seed AI vs Cyc, where does Novamente fall? > > > Ok, let's get rolling then. > > Ben, here's a question. To what extent are parts of Novamente > hand built ala Cyc? > > I can easily imagine a dimension here. At one end is Cyc, which > is carefully and meticulously constructed by people. The design > work is two fold, creating the structure within which to store > information, and then the inputting of the information. At the > other end of the dimension is pure Seed AI. A know-nothing baby. > You put it together, hit the power button and let it > grow(admittedly it would require tuning as it matured). > > Where would Novamente fall along this dimension? > > What domains of knowledge do you expect to have to explicitly > create by hand? > > -Brad > > > ------- > To unsubscribe, change your address, or temporarily deactivate > your subscription, > please go to http://v2.listbox.com/member/[EMAIL PROTECTED] > ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
