3 ms·
Another less know but really promising approach is program synthesis (also called "program generation"). One can build fairly robust model just using 2-5 exampl
by sytelus 10y ago
Another less know but really promising approach is program synthesis (also called "program generation"). One can build fairly robust model just using 2-5 examples and that too in just seconds. Implementation of this approach is already shipped in to Excel where you just enter few example of formatting and "Flash Fill" will learn what to do: http://research.microsoft.com/en-us/um/people/sumitg/flashfill.html http://research.microsoft.com/en-us/um/people/sumitg/flashfi...
Paper: http://research.microsoft.com/en-us/um/people/sumitg/pubs/cacm12-synthesis.pdf http://research.microsoft.com/en-us/um/people/sumitg/pubs/ca...
- prats226 10y agoThis is cool. From what I understand from paper, its DSL has set of algorithms as building blocks that learn the input/output function. Deep learning algos are trying to do the same but with more generic blocks where assumption is that a lot of these blocks will be able to learn algorithms too. Deep learning is trying to build with a more generic approach in which transfer learning is helping to reduce number of examples needed by reusing algorithms learned.