user avatar
user avatar
user avatar
OmG
  • Member for 5 years, 7 months
  • Last seen this week
757 votes
11 answers
953k views

How to choose the number of hidden layers and nodes in a feedforward neural network?

  • 52.9k
347 votes
5 answers
372k views

What is the trade-off between batch size and number of iterations to train a neural network?

210 votes
8 answers
239k views

What are the advantages of ReLU over sigmoid function in deep neural networks?

  • 11.7k
92 votes
10 answers
202k views

How should outliers be dealt with in linear regression analysis?

  • 4,196
71 votes
3 answers
96k views

What's the difference between feed-forward and recurrent neural networks?

  • 12k
55 votes
6 answers
7k views

What are the main theorems in Machine (Deep) Learning?

53 votes
4 answers
42k views

What is perplexity?

  • 4,107
36 votes
1 answer
35k views

Why KL divergence is non-negative?

30 votes
2 answers
18k views

What is "reduced-rank regression" all about?

  • 8,025
18 votes
3 answers
20k views

How to uniformly project a hash to a fixed number of buckets

  • 1,232
10 votes
1 answer
3k views

Posterior variance vs variance of the posterior mean

  • 6,338
8 votes
1 answer
253 views

Do random variables follow the same algebraic rules as ordinary numbers?

8 votes
2 answers
1k views

What is the problem with $p > n$?

  • 137
7 votes
2 answers
5k views

Is PCG random number generator as good as claimed?

  • 173
4 votes
1 answer
767 views

Hidden Markov model with reward

  • 101
4 votes
1 answer
145 views

Workaround for word embeddings that do not "see" antonyms

  • 51
4 votes
1 answer
107 views

Variance inequality

  • 115
3 votes
1 answer
352 views

Dimension reduction methods: overview of complexity

2 votes
0 answers
769 views

PCA: how to select eigen vectors corresponding to small eigenvalues for regression

  • 1,319
1 vote
1 answer
1k views

Scott's and Freedman–Diaconis rules of the thumb for selecting bin width - disatvantages

1 vote
0 answers
43 views

Question regarding the Supervised Principal Components method

  • 11