Questions asked by Alexandre Passos - MetaOptimize Q+A

Questions asked by Alexandre Passos

active newest hottest most voted

1

votes

1

answers

3k

views

What are the main advantages and disadvantages of dual decomposition versus A* (and search-based methods) for NLP applications?

Oct 12 '10 at 19:21 Alexandre Passos ♦23061

a-star nlp search dual-decomposition

9

votes

3

answers

4k

views

Why are coordinate ascent algorithms (EM, variational EM) preferred over gradient methods for generative probabilistic models?

Aug 11 '10 at 06:29 Alexandre Passos ♦23061

bayesian em newton optimization variational-inference

4

votes

3

answers

2k

views

Which language/framework do you use to implement (directed) graphical models?

Jul 31 '10 at 00:48 yoavg726 (asked by Alexandre Passos ♦23061 )

libraries graphical-models mcmc language variational-inference

2

votes

2

answers

1k

views

In learning with kernels, are RKHS only used to allow the use of a mercer kernel or is there some other essential property?

Jul 16 '10 at 12:58 Alexandre Passos ♦23061

kernel regularization svm rkhs feature learning-theory

5

votes

3

answers

2k

views

What is a good reference for understanding surrogate loss functions?

Jul 13 '10 at 20:34 Mark Reid296 (asked by Alexandre Passos ♦23061 )

supervised-learning loss-function optimization classification learning-theory

6

votes

5

answers

1k

views

Collaboration and open problems in Machine Learning and related fields

Jul 09 '10 at 07:20 zeno211 (asked by Alexandre Passos ♦23061 )

ml collaboration novice open-problem

1

votes

1

answers

1k

views

"Vanishing" classes in gibbs samplers for dirichlet-multinomial models

Jul 01 '10 at 00:04 aria422009 (asked by Alexandre Passos ♦23061 )

hyperparameters dirichlet-multinomial tuning gibbs sampling

3

votes

2

answers

2k

views

Is there a good reference for blocked gibbs sampling?

Jun 30 '10 at 13:23 Alexandre Passos ♦23061

bayesian blocked gibbs sampling

38Questions asked by Alexandre Passos

91answers

Newest questions are shown first.

Tags

classification × 164
nlp × 147
svm × 98
optimization × 90
bayesian × 62
sampling × 49
graphical-models × 36
search × 30
regularization × 27
kernel × 26
mcmc × 21
supervised-learning × 18
em × 17
learning-theory × 16
variational-inference × 14
feature × 13
language × 11
gibbs × 10
hyperparameters × 10
libraries × 10
loss-function × 10
ml × 10
dirichlet-multinomial × 4
rkhs × 4
tuning × 4
newton × 3
blocked × 2
dual-decomposition × 2
a-star × 1
collaboration × 1
novice × 1
open-problem × 1

posts per page103050

learn more | privacy | contact

powered by OSQA

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.