haarts - user overview - MetaOptimize Q+A

haarts

96

karma

Registered user
member since	Jul 04 '13 at 06:24
last seen	Jul 18 '13 at 08:33

overview recent activity karma history favorites

3 Questions

1

votes

2

answers

2k

views

Why is a RBM able to be generative and an Autoencoder not?

Jul 18 '13 at 17:31 Alexandre Passos ♦25541●54●278●421 (asked by haarts96●3●5●5 )

generative rbm autoencoder

1

votes

0

answers

1k

views

Denoising Autoencoder not performing on CIFAR10

Nov 15 '14 at 19:58 Saul Berardo66●1●2●7 (asked by haarts96●3●5●5 )

denoising rbm autoencoder training cifar10

4

votes

1

answers

960

views

Why should the reconstruction loss of an auto encoder be HIGH on non training examples?

Jul 04 '13 at 11:07 Alexandre Passos ♦25541●54●278●421 (asked by haarts96●3●5●5 )

loss autoencoder reconstruction

0 Answers

0 Votes

0

0

8 Tags

autoencoder × 4
rbm × 3
generative × 2
loss × 1
training × 1
cifar10 × 1
denoising × 1
reconstruction × 1

7 Kudos

● Prominent Question × 3
● Well-Known Question × 3
● Conspicuous Question × 3
● Inquisitive × 1
● Prized Question × 1
● Appreciated Question × 1

● Esteemed Question × 1

learn more | privacy | contact

powered by OSQA

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