Mercurial > pylearn

comparison doc/v2_planning/optimization.txt @ 1027:a1b6ccd5b6dc

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
few comments added
author Yoshua Bengio <bengioy@iro.umontreal.ca>
date Fri, 03 Sep 2010 21:57:22 -0400
parents 618b9fdbfda5
children 89e76e6e074f
comparison
equal deleted inserted replaced
1026:38f799f8b6cd 1027:a1b6ccd5b6dc
28 - sgd 28 - sgd
29 - sgd + momentum 29 - sgd + momentum
30 - sgd with annealing schedule 30 - sgd with annealing schedule
31 - TONGA 31 - TONGA
32 - James Marten's Hessian-free 32 - James Marten's Hessian-free
33 - Conjugate gradients, batch and (large) mini-batch [that is also what Marten's thing does]
33 34
34 Do we need anything to make batch algos work better with Pylearn things? 35 Do we need anything to make batch algos work better with Pylearn things?
35 - conjugate methods? 36 - conjugate methods? yes
36 - L-BFGS? 37 - L-BFGS? maybe, when needed
37 38
39
40
41