Sander Dieleman

1,556

karma

Registered user
real name	Sander Dieleman
member since	Dec 26 '11 at 08:08
last seen	Oct 16 '14 at 09:01
user's website	http://benanne.github.io/
location	Belgium

PhD student at Ghent University

overview recent activity karma history favorites

6 Questions

votes

answers

views

Sampling dropout masks for minibatch SGD

Apr 02 '13 at 18:59 gdahl ♦3414●5●35●59 (asked by Sander Dieleman1556●7●27●34 )

votes

answers

views

Effect of parameter initialisation in RBM training

Jun 27 '12 at 08:22 Sander Dieleman1556●7●27●34

deep-learning training rbm initialisation

votes

answers

views

Symmetric energy functions for RBMs

Mar 19 '12 at 10:37 Sander Dieleman1556●7●27●34

symmetry energy-functions energy precision rbm variance products-of-experts

votes

answers

views

What is the advantage of learning variance in Gaussian RBMs

Jun 06 '12 at 19:11 gdahl ♦3414●5●35●59 (asked by Sander Dieleman1556●7●27●34 )

contrastive divergence gaussian rbm units variance

votes

answers

views

Energy function of an RBM with noisy rectified linear units (NReLUs)

Feb 13 '13 at 15:09 Dan Ryan406●7●11●16 (asked by Sander Dieleman1556●7●27●34 )

boltzmann linear energy-functions restricted energy rectified machine nrelu rbm units

votes

answers

views

Mean field approximation in RBM training

Dec 26 '11 at 13:07 Sander Dieleman1556●7●27●34

approximation rbm mean-field cd

29 Answers

How to compute the free energy of a RBM given its energy?

Deep learning and rectified linear units

Connectivity between subsequent feature maps in CNNs

How to understand learning the identity function in auto-encoder as mentioned in the paper "Learning Deep Architectures for AI"?

Addition of matrices with un-matched dimensions

Dropout in convolutional layer?

Issue reconstructing input images in Convolutional Sparse Autoencoders

Relation between "filter" and "feature map" in convolutional neural networks

non-linear NCA ovefits even with the regularizers ,, what can I try?

What types of unsupervised feature extraction are out there?

Dropout: how to scale up weights after pretraining when the related 2 layers has different dropout probability?

Predictive Sparse Decomposition vs Sparse AutoEncoders?

Why Theano lacks of convolution with border_mode='same'?

DeepLearningTutorials: How to integrate color information?

Autoencoder with linear decoding layer vs non-linear decoding layer

What is the efficiency of theano based codes comparing to C/C++ implemented codes?

sampling visible units in RBM

Monitoring the training of Restricted Boltzmann Machines

deep learning (layer\hierarchy features visualization)

Problems with DBN for audio data

Pretraining Neural Networks: Autoencoders vs Restricted Boltzmann Machines

Effect of parameter initialisation in RBM training

Book recommendation on recommender systems?

Zero component analysis

learning the vector representation

Meaning of x.x^T

RBM linear visible units

Replicated k-hot RBM units

theano and adaptive learning rate in training MLP

130 Votes

130

units × 4
energy × 4
dropout × 4
mnist × 3
variance × 3
machine × 2
linear × 2
mean-field × 2
rectified × 2
gaussian × 2

rectified-linear-units × 2
vector × 2
cd × 2
restricted × 2
python × 2
divergence × 2
prediction × 2
unsupervised-learning × 2
conv2d × 2
boltzmann × 2

nrelu × 2
approximation × 2
parameters × 2
convolutional × 2
contrastive × 2
color × 1
feature-learning × 1
training × 1
symmetry × 1
similarity × 1

minibatch × 1
pylearn × 1
forecasting × 1
diagrams × 1
system × 1
generalization × 1
convolution-neural-networks × 1
recommender × 1
topic-models × 1
nearest × 1

22 Kudos

● Appreciated Answer × 16
● Cognoscenti × 9
● Conspicuous Question × 6
● Well-Known Question × 6
● Prominent Question × 6
● Esteemed Answer × 5

● Appreciated Question × 3
● Esteemed Question × 2
● Prized Question × 2
● Commentator × 1
● Participant × 1
● Adulated Answer × 1

● Guide × 1
● Auto-Didact × 1
● Blue Pencil × 1
● Mentor × 1
● Advocate × 1
● Cherished Question × 1

● Revisitation × 1
● Revisionist × 1
● Vigilant × 1
● On the record × 1

Registered user

6 Questions

Sampling dropout masks for minibatch SGD

Effect of parameter initialisation in RBM training

Symmetric energy functions for RBMs

What is the advantage of learning variance in Gaussian RBMs

Energy function of an RBM with noisy rectified linear units (NReLUs)

Mean field approximation in RBM training

29 Answers

130 Votes

50 Tags

22 Kudos