Brian Vandenberg

824

karma

Registered user
real name	Brian Vandenberg
member since	Dec 01 '10 at 00:43
last seen	Mar 18 at 11:18
location	Utah, USA

I am not currently a student. My background is in CS & math.

overview recent activity karma history favorites

19 Questions

votes

answers

565

views

Methods for training kernel-based models on large data-sets

Apr 10 '12 at 10:11 Brian Vandenberg824●21●37●46

kernel-trick large-data-sets

votes

answers

views

Value of deep -vs- shallow NNs

Mar 18 '13 at 16:46 Dan Ryan406●7●11●16 (asked by Brian Vandenberg824●21●37●46 )

deep-learning

votes

answers

views

Applying the R{} operator to RBMs; how to treat the function for sampling from a distribution?

Mar 09 '12 at 11:19 Brian Vandenberg824●21●37●46

r-operator rbm hessian-free

votes

answers

views

Whether to use raw output of transfer functions -vs- 'sampled' values, and why

Feb 22 '12 at 22:08 Brian Vandenberg824●21●37●46

noise sampling rbm rectified-linear-units neural-networks latent-variables

votes

answers

views

Current state of the art in image segmentation

Dec 11 '11 at 05:28 Andreas Mueller2686●18●58●93 (asked by Brian Vandenberg824●21●37●46 )

state-of-the-art image-segmentation

votes

answers

views

Filtering near-duplicate samples

Oct 15 '11 at 02:11 Mathieu Blondel1191●2●16●15 (asked by Brian Vandenberg824●21●37●46 )

duplicate-removal samples

votes

answers

views

Strategies for embedding a means for identifying an algorithm

Aug 06 '11 at 10:46 alto603●5●11●24 (asked by Brian Vandenberg824●21●37●46 )

signatures algorithms

votes

answers

views

Is comparing points of interest in two separate images to determine similarity an NP or NP hard problem?

Jul 12 '11 at 03:43 Alexandre Passos ♦25541●54●278●421 (asked by Brian Vandenberg824●21●37●46 )

votes

answers

views

[meta] karma not updating anymore

Jul 13 '11 at 14:06 Joseph Turian ♦♦5790●51●125●146 (asked by Brian Vandenberg824●21●37●46 )

meta karma

votes

answers

views

hessian-free RNN questions

Dec 21 '13 at 11:38 Alex Vlasov61●1●2●7 (asked by Brian Vandenberg824●21●37●46 )

hessian-free paper-understanding rnn recurrent-neural-networks

votes

answers

views

Deriving RNN backprop equations

Jun 08 '11 at 02:21 Justin Bayer1706●9●30●45 (asked by Brian Vandenberg824●21●37●46 )

rnn recurrent-neural-networks

votes

answers

832

views

[meta] Nested lists not rendering properly

Apr 08 '11 at 16:47 Brian Vandenberg824●21●37●46

Question re: Gauss-newton matrix vector product

Mar 19 '11 at 08:11 Philemon Brakel2445●10●35●60 (asked by Brian Vandenberg824●21●37●46 )

hessian-free gauss-newton

votes

answers

views

Question re: paper on 'fast multiplication by the hessian'

Jan 04 '13 at 16:43 Brian Vandenberg824●21●37●46

paper-understanding hessian-free

votes

answers

views

Unsupervised methods for learning [artificial] support vectors for kernel methods?

Jun 05 '11 at 02:46 Jacob Jensen1914●31●56●63 (asked by Brian Vandenberg824●21●37●46 )

kernel unsupervised kernels kernel-trick kernel-methods unsupervised-learning

View all questions by Brian Vandenberg

57 Answers

Conditional RBM / RNN vs. HMM in handling latencies

Gibbs sampling parameters of Normal-InverseWishart distribution, does NIW have a prior itself?

Role of transfer functions in neural network

Why is AI considered to be part of computer science?

What's the difference between conditional RBMs and recurrent neural networks?

SIFT can ignore black backgrounds?

Data at multiple scales

Measuring quality of unsupervised feature extraction

Dataset wanted: recent chat log archives of a technical nature?

Good Freely Available Videos on Machine Learning

log partition function

Where to find good research papers

Is it worth avoiding intermediate matrices in calculations?

Sumatory manipulation

Application of machine learning to fMRI/EEG data?

hessian-free RNN questions

How are new models invented?

Tricks for handling big data: What to do when you have more data than CPU to crunch it

What are some good resources for learning about large-scale machine learning and optimization?

"Backtracking step" in Exponentiated Gradient Descent

Are there any requirements for L1 regularization?

What is dot product intuitively?

about Support Vector Machines and Kernel methods

What are some Novel, Emerging Areas of Machine Learning

hi, what's these codes used for in matlab topic modelling toolbox?

LDA is slow, so is it worth researching?

Estimating class-cond distributions for bag-of-word features in Naive Bayes

Why do people use negative log scores, instead of just log scores?

When to use L1 regularization and when L2?

View all questions answered by Brian Vandenberg

129 Votes

129

conjugate-gradient × 12
machine-learning × 11
kernel-trick × 8
state-of-the-art × 6
unsupervised-learning × 6
image-segmentation × 6
unsupervised × 6
optimization × 6
nlp × 5
kernels × 5

mathematics × 5
kernel-methods × 5
kernel × 5
r-operator × 4
bolzmann-machines × 4
probability × 4
algorithms × 4
data × 4
energy-functions × 4
distributions × 4

tutorial × 3
big-picture × 3
sift × 3
time-series × 3
large-data-sets × 3
distributed × 3
regularization × 3
feature-extraction × 3
signatures × 3
video × 3

math × 3
papers × 3
markov-random-field × 3
meta × 3
dbn × 3
svm × 2
icml × 2
boltzmann-machines × 2
np-hard × 2
log-linear × 2

22 Kudos

● Prominent Question × 19
● Well-Known Question × 19
● Conspicuous Question × 19
● Appreciated Answer × 13
● Appreciated Question × 6
● Esteemed Answer × 5

● Prized Question × 4
● Esteemed Question × 3
● Revisitation × 2
● Cognoscenti × 2
● Participant × 1
● Adulated Answer × 1

● Revisionist × 1
● Blue Pencil × 1
● Mentor × 1
● Advocate × 1
● Inquisitive × 1
● Cherished Question × 1

● Guide × 1
● Adulated Question × 1
● Pronouncement × 1
● Commentator × 1

Registered user

19 Questions

57 Answers

129 Votes

50 Tags

22 Kudos