User Felipe

29 votes

4 answers

23k views

Does the image format (png, jpg, gif) affect how an image recognition neural net is trained?

asked Jun 17, 2017 at 22:54

14 votes

2 answers

24k views

Simple Linear Regression in Keras

asked Jun 8, 2017 at 5:40

7 votes

4 answers

7k views

What is the best form (Gaussian, Multinomial) of Naive Bayes to use with categorical (one-hot encoded) features?

asked May 29, 2016 at 4:32

4 votes

1 answer

161 views

Difference between contrastive divergence and importance sampling

asked Feb 19, 2017 at 3:21

4 votes

1 answer

3k views

NNs: Multiple Sigmoid + Binary Cross Entropy giving better results than Softmax + Categorical Cross Entropy

asked Jun 14, 2017 at 5:49

3 votes

1 answer

314 views

Why does standardizing my data cause better results? [duplicate]

asked Mar 13, 2017 at 5:05

3 votes

1 answer

41 views

Does adding random null values to features at train time make models less 'sensitive' to those?

asked Oct 20 at 5:08

1 vote

1 answer

165 views

Same exact model: converges with adagrad, diverges with adadelta

asked Feb 28, 2018 at 2:38

1 vote

1 answer

490 views

Attention Mechanisms and Alignment Models in Machine Translation

asked Jun 28, 2020 at 4:36

0 votes

1 answer

2k views

Difference between "transfer learning" and "fine-tuning"

asked Jan 22, 2023 at 21:23

0 votes

1 answer

34 views

Self-supervised Target Definition in the Original Neural Language Model by Bengio et al (2003)

asked Feb 22, 2023 at 3:26

Stack Exchange Network

11 Questions

Does the image format (png, jpg, gif) affect how an image recognition neural net is trained?

Simple Linear Regression in Keras

What is the best form (Gaussian, Multinomial) of Naive Bayes to use with categorical (one-hot encoded) features?

Difference between contrastive divergence and importance sampling

NNs: Multiple Sigmoid + Binary Cross Entropy giving better results than Softmax + Categorical Cross Entropy

Why does standardizing my data cause better results? [duplicate]

Does adding random null values to features at train time make models less 'sensitive' to those?

Same exact model: converges with adagrad, diverges with adadelta

Attention Mechanisms and Alignment Models in Machine Translation

Difference between "transfer learning" and "fine-tuning"

Self-supervised Target Definition in the Original Neural Language Model by Bengio et al (2003)