As a related note, I think it is also possible to learn parse trees (hence the tree data structure) using reinforcement learning for images and natural language,

For images:
http://vision.mas.ecp.fr/Personnel/teboul/files/cvpr11_teboul.pdf

For natural language processing, I think Hal Daume published a few papers with imitation learning.

But these examples are not using neural networks.



On Thu, Jan 16, 2014 at 2:09 PM, Yoshua Bengio <yoshua.bengio@gmail.com> wrote:
Here is the paper I mentioned during Ian's presentation about training a recurrent net to exploit a push-pop stack:

http://books.nips.cc/papers/files/nips02/0380.pdf

It was not Mike Mozer but Lee Giles.

-- Yoshua



On Mon, Jan 13, 2014 at 1:04 PM, Razvan Pascanu <r.pascanu@gmail.com> wrote:
Hi all,

 First of all the schedule has changed, and tea talks will be Wednesdays from 13-14 from now on.

There is an exception this week. We will have Ian presenting this  **Thursday from 13:00 to 14:00**.
Speaker: Ian Goodfellow

Title: Ian's ideas for research projects

Abstract:

I'll throw out a few of my recent ideas for research
projects, ranging from easy to idealistic:
-Restricted maxout units
-Regularizing piecewise linear nets to change pieces rarely
-Sparsely connected recurrent nets
-Trajectory optimization for recurrent nets
-"Cognitive agency" and how you can use it to do things like
discrete-state nets, dynamically structured nets, and nets that
interact with data structures like stacks and tapes

I hope to see many of you there. And please register for giving tea talks.

Razvan

_______________________________________________
Lisa_labo mailing list
Lisa_labo@iro.umontreal.ca
https://webmail.iro.umontreal.ca/mailman/listinfo/lisa_labo



_______________________________________________
Lisa_labo mailing list
Lisa_labo@iro.umontreal.ca
https://webmail.iro.umontreal.ca/mailman/listinfo/lisa_labo




--
Caglar GULCEHRE