user avatar
user avatar
user avatar
meTchaikovsky
  • Member for 4 years, 3 months
  • Last seen this week
36 votes
1 answer
34k views

Why KL divergence is non-negative?

14 votes
4 answers
7k views

What causes lasso to be unstable for feature selection?

8 votes
1 answer
403 views

Sparse linear regression 0-norm and 1-norm

7 votes
2 answers
5k views

Why does larger perplexity tend to produce clearer clusters in t-SNE?

5 votes
1 answer
200 views

Linear regression with $l_0$ regularization

5 votes
1 answer
102 views

A question about the inversion method

4 votes
2 answers
2k views

KL divergence invariant to affine transformation?

4 votes
1 answer
510 views

How to continuously rotate a bivariate normal distribution by changing its correlation matrix?

3 votes
1 answer
374 views

Train an ensemble of neural networks on different datasets, what is the best way to scale the inputs?

3 votes
0 answers
29 views

Increase the number of samples when the PDF is invariant

3 votes
1 answer
164 views

What can we learn from this visualization?

3 votes
1 answer
588 views

Variational autoencoder obtain high likelihood but produce low quality sample?

2 votes
0 answers
318 views

How do we train variational autoencoders in practice?

2 votes
1 answer
349 views

Feature filtering with LASSO and cross validation

2 votes
1 answer
834 views

What is $\lambda_{min}$ that makes all regression coefficients zero in LASSO regression [duplicate]

2 votes
0 answers
63 views

Select 3 features from 4000 correlated features

2 votes
0 answers
1k views

Prove linear transformation of gaussian random vectors are still gaussian random vectors

2 votes
1 answer
2k views

Data sparsity becomes a problem

2 votes
1 answer
569 views

Can we still use conjugate gradient descent when we don't know the Hessian matrix?

1 vote
1 answer
49 views

How to differentiate through the optimization process in Model-Agnostic Meta-Learning?

1 vote
1 answer
1k views

Can t-SNE help feature selection?

1 vote
0 answers
55 views

Why it is hard for `xgboost` to learn periodic functions?

1 vote
1 answer
64 views

Should I discard a feature whose max - min is smaller than 1e-5?

1 vote
1 answer
425 views

Joint distribution in latent Dirichlet model

1 vote
1 answer
93 views

$\lambda \Vert k \Vert_0$ or $\Vert k \Vert_0 \leqslant n$

0 votes
2 answers
331 views

Validation set for early stopping

0 votes
0 answers
86 views

How SVD is used for dimensionality reduction? [duplicate]

0 votes
1 answer
976 views

Maximum likelihood estimate for a univariate gaussian

0 votes
1 answer
47 views

Discard features with small variance, how to do in practice? [duplicate]

0 votes
0 answers
206 views

Why does `xgboost` find such an unbalanced split in my data?