Neural Turing machine

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Citation bot (talk | contribs) at 04:19, 13 July 2016 (Add: year, class, eprint, author pars. 1-3. Removed parameters. You can use this bot yourself. Report bugs here.). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Neural turing machines combine the fuzzy pattern matching capabilities of neural networks with the algorithmic power of programmable computers. They are an extension of neural networks that are coupled to external memory resources, which they interact with through attentional mechanisms. Their memory interactions are differentiable end-to-end, making it possible to optimize them using gradient descent.[1] They can infer simple algorithms such as copying, sorting, and associative recall from input and output examples.[2]

They can infer algorithms from input and output examples alone.

References

  1. ^ "Deep Minds: An Interview with Google's Alex Graves & Koray Kavukcuoglu". Retrieved May 17, 2016.
  2. ^ Graves, Alex; Wayne, Greg; Danihelka, Ivo (2014). "Neural Turing Machines". arXiv:1410.5401 [cs.NE]. {{cite arXiv}}: Cite has empty unknown parameter: |newspaper= (help)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Neural_Turing_machine&oldid=729574381"