User user56834

22 votes

4 answers

3k views

Textbook on the theory of neural nets/ML algorithms?

asked Apr 29, 2018 at 13:22

17 votes

1 answer

5k views

Is the mean (Bayesian) posterior estimate of $\theta$ a (Frequentist) unbiased estimator of $\theta$?

asked Nov 12, 2017 at 14:03

13 votes

2 answers

7k views

Interpretation of entropy for continuous distribution?

asked May 21, 2018 at 14:50

12 votes

1 answer

2k views

Interpretation of Radon-Nikodym derivative between probability measures?

asked Feb 1, 2018 at 5:29

12 votes

3 answers

3k views

Is the linearity assumption in linear regression merely a definition of $\epsilon$?

asked Jan 2, 2018 at 12:19

10 votes

2 answers

2k views

Is the invariance property of the ML estimator nonsensical from a Bayesian perspective?

asked Nov 14, 2017 at 8:35

9 votes

1 answer

1k views

Why can't we cancel these two matrices in the OLS estimator?

asked Oct 25, 2017 at 14:59

9 votes

1 answer

2k views

Interpretation of "score"

asked Jan 31, 2018 at 17:17

9 votes

3 answers

1k views

How can we interpret a neural network with sgd from a Bayesian perspective?

asked Jun 21, 2018 at 7:32

8 votes

3 answers

926 views

Introduction to frequentist statistics for Bayesians [closed]

asked Nov 14, 2017 at 6:59

8 votes

2 answers

789 views

What is the justification of using taylor approximations inside expectation operators?

asked Dec 31, 2017 at 13:46

7 votes

1 answer

260 views

What is the name of $E(XY)$? [duplicate]

asked May 18, 2018 at 4:32

7 votes

1 answer

547 views

Is there a "central distribution" for distributions for which the CLT doesn't apply?

asked Feb 10, 2018 at 15:48

7 votes

2 answers

877 views

Are loss functions necessarily additive in observations?

asked Jun 7, 2019 at 9:21

6 votes

2 answers

142 views

Theoretical justification for minimizing $KL(q_\phi|p)$ rather than $KL(p|q_\phi)$?

asked Jun 15 at 14:35

6 votes

2 answers

1k views

Why is the bayesian information criterion called that way?

asked Jun 6, 2019 at 14:01

6 votes

1 answer

6k views

Why does this neural network in keras fail so badly?

asked Aug 17, 2016 at 15:12

5 votes

1 answer

3k views

How do we derive that $S^2$ is chi-squared distributed (with $n-1$ df)?

asked Nov 12, 2017 at 15:47

5 votes

2 answers

1k views

What's the difference between "Optimal linear predictor" and "best unbiased linear estimator"?

asked Jan 3, 2018 at 13:44

5 votes

2 answers

5k views

"Law of large numbers" for distribution with infinite variance?

asked Feb 11, 2018 at 14:05

5 votes

0 answers

1k views

Partial correlation vs conditional correlation? [duplicate]

asked Feb 18, 2018 at 19:04

5 votes

2 answers

15k views

When should an F-test be one sided vs two sided? [duplicate]

asked May 16, 2018 at 15:28

5 votes

2 answers

253 views

Term for "extent to which a test throws away information"?

asked Jan 11, 2018 at 19:06

5 votes

2 answers

2k views

Proving OLS unbiasedness without conditional zero error expectation?

asked Jan 18, 2018 at 9:10

5 votes

1 answer

2k views

Why don’t we need strict exogeneity for OLS consistency? [closed]

asked Jun 2, 2018 at 9:38

5 votes

1 answer

253 views

Proof for dimension of chi-squared test?

asked Jun 7, 2018 at 4:33

5 votes

1 answer

1k views

Difference between tight and uniformly tight random variables?

asked Jun 10, 2019 at 9:50

5 votes

3 answers

299 views

Why intuitively does $\mathbb E(\frac d {d\theta}\log p_\theta(x))=0$?

asked Jun 23, 2019 at 13:44

4 votes

1 answer

176 views

Topological rather than metric based machine learning theory?

asked Jun 4, 2019 at 7:04

4 votes

2 answers

110 views

Learning from multiple very varied data sets?

asked Nov 9, 2019 at 8:36

Stack Exchange Network

88 Questions

Textbook on the theory of neural nets/ML algorithms?

Is the mean (Bayesian) posterior estimate of $\theta$ a (Frequentist) unbiased estimator of $\theta$?

Interpretation of entropy for continuous distribution?

Interpretation of Radon-Nikodym derivative between probability measures?

Is the linearity assumption in linear regression merely a definition of $\epsilon$?

Is the invariance property of the ML estimator nonsensical from a Bayesian perspective?

Why can't we cancel these two matrices in the OLS estimator?

Interpretation of "score"

How can we interpret a neural network with sgd from a Bayesian perspective?

Introduction to frequentist statistics for Bayesians [closed]

What is the justification of using taylor approximations inside expectation operators?

What is the name of $E(XY)$? [duplicate]

Is there a "central distribution" for distributions for which the CLT doesn't apply?

Are loss functions necessarily additive in observations?

Theoretical justification for minimizing $KL(q_\phi|p)$ rather than $KL(p|q_\phi)$?

Why is the bayesian information criterion called that way?

Why does this neural network in keras fail so badly?

How do we derive that $S^2$ is chi-squared distributed (with $n-1$ df)?

What's the difference between "Optimal linear predictor" and "best unbiased linear estimator"?

"Law of large numbers" for distribution with infinite variance?

Partial correlation vs conditional correlation? [duplicate]

When should an F-test be one sided vs two sided? [duplicate]

Term for "extent to which a test throws away information"?

Proving OLS unbiasedness without conditional zero error expectation?

Why don’t we need strict exogeneity for OLS consistency? [closed]

Proof for dimension of chi-squared test?

Difference between tight and uniformly tight random variables?

Why intuitively does $\mathbb E(\frac d {d\theta}\log p_\theta(x))=0$?

Topological rather than metric based machine learning theory?

Learning from multiple very varied data sets?