Cross Validated
Sponsored by
Sponsored logo
Hugh Perkins
Meta user
Network profile
Summary Answers Questions Tags Badges Favorites Bounties Reputation All actions

106 Answers

Newest Activity Votes
14
Gradient backpropagation through ResNet skip connections
Mar 21 '17 at 8:42
13
Can a standard deviation of raw scores be reported as a standard deviation of percentages?
Oct 4 '17 at 10:49
12
What are “residual connections” in RNNs?
Jan 1 '18 at 21:01
9
How to train LSTM model on multiple time series data?
Jul 25 '18 at 1:56
9
Data augmentation on training set only?
Dec 31 '17 at 3:00
9
What is the difference between regular PCA and probabilistic PCA?
Oct 9 '17 at 20:56
8
need intuition: why is there no winning strategy to this game?
Mar 26 '17 at 21:01
8
What is the root cause of the class imbalance problem?
Mar 18 '17 at 23:19
7
How to interpret the results when both ridge and lasso separately perform well but produce different coefficients
Mar 14 '17 at 12:44
6
Modern Use Cases of Restricted Boltzmann Machines (RBM's)?
Oct 9 '17 at 20:46
6
How do we incorporate new information into a Dirichlet prior distribution?
Oct 9 '17 at 14:17
5
What is the reason that the Adam Optimizer is considered robust to the value of its hyper parameters?
Oct 9 '17 at 20:35
5
Interpret predictions of black box models
Mar 27 '17 at 12:36
5
LSTM clarification on output
Mar 26 '17 at 13:24
4
When is $\mathbb{E}\left[\frac{1}{\sum X_{i}}\right] = \frac{1}{\mathbb{E}\left[\sum X_{i}\right]}$?
Oct 16 '17 at 8:01
4
In the Metropolis algorithm, if the ratio of probabilities $r$ is less than $1$, why not directly reject instead of accepting with probability $r$?
Mar 23 '17 at 23:51
3
Convolutional neural networks backpropagation
Feb 11 '18 at 23:59
3
For symmetric matrices, is the Cholesky decomposition better than the SVD?
Feb 11 '18 at 16:45
3
Sequence to Sequence models: Why is the start sequence (e.g. “</s>”) required?
Jan 2 '18 at 2:34
3
If z is uniform normal is $f(z;\theta)$ a normal distribution
Nov 25 '17 at 15:45
3
How to train a CNN with multiple heads?
Mar 24 '17 at 12:10
3
MCMC: invalid covariance matrix due to numerical error
Apr 2 '17 at 7:44
3
How to distinguish episodic task and continuous tasks?
Apr 2 '17 at 8:50
3
Assumptions on class priors in expectation-maximization?
Mar 27 '17 at 8:55
2
Is there a known name / keyword to describe continuous manual improvement of the labels used to train a classifier?
Jul 28 '18 at 0:14
2
Deep neural networks versus tall neural networks
Jul 24 '18 at 0:11
2
How to update kernel values in a convolutional layer during backward pass?
Jul 24 '18 at 0:08
2
Stateful LSTM for time-series prediction - should each input sequence be shifted by 1 time step or sequenceLength time steps
Jul 9 '18 at 3:07
2
Do Autoencoders preserve distances?
Jun 15 '18 at 22:29
2
deep learning - word embedding with parts of speech
Feb 12 '18 at 0:08
1 2 3 4