Transfer Learning

From OpenCog
Jump to: navigation, search

Many papers on transfer learning are listed here

http://www.cse.ust.hk/~sinnopan/conferenceTL.htm

Another one was in the AGI-08 proceedings.

I (Ben G) think that this paper

http://scholar.google.com/scholar?hl=en&lr=&safe=off&cluster=2777703100166145782

is an interesting idea that could apply to PLN as well as to (what they talk about in the paper) MLN ....

What it suggests is an inference control mechanism involving searching for homomorphisms, and then using these for probabilistic inference guidance...

Two two somewhat related papers suggested by Lukasz Stafiniak:

Graph-Based Domain Mapping for Transfer Learning in General Games. Gregory Kuhlmann, Peter Stone http://www.cs.utexas.edu/~pstone/Papers/bib2html/b2hd-ECML07-rulegraphs.html

Intrinsically Motivated Reinforcement Learning. Nuttapong Chentanez Andrew G. Barto Satinder Singh http://www.eecs.umich.edu/~baveja/Papers/FinalNIPSIMRL.pdf