I wanted to know what researchers in theoretical computer science, machine learning, math and AI think of a software system which learns by itself.
Can such a system be called a self-aware system? Meaning a system which is aware of itself. And its 'self' is knowledge.
The application is something like knowledge graph in google. It can be better than that. People can query the system to tell something about itself, which will be knowledge of our world.
If anybody have pointers regarding this topic, please guide me on this.
Hi. Systems that learn does not necassery need to be self-ware. However, there is research into autonomous systems that is investigating something called self-* where * refers to different things, like self-awareness, self-defence and so on. See for instance: http://www.cs.unibo.it/~babaoglu/papers/pdf/Self-Star%20Vision.pdf
Hi,
if I understood your question correctly, the research behind IBM's Watson system (see linked article) could be relevant for you. You can also check out the later developments: http://www.informatik.uni-trier.de/~ley/pers/hd/f/Ferrucci:David_A=
Cheers,
Zoltan
http://www.aaai.org/Magazine/Watson/watson.php
Tomas, Thanks a ton. That article helped me a lot. :)
Thanks everyone. I got an initial idea of how to proceed further.
Preprint Intellectual activity, knowledge, information, data... An at...
- Badges
- Science topic