The trend suggests that as you increase the width of a neural network, the accuracy increases till a certain threshold value, and then starts decreasing. What could be the possible reason for this decrease

Question

miloubbb1533 miloubbb1533

03-11-2020
Mathematics

contestada

The trend suggests that as you increase the width of a neural network, the accuracy increases till a certain threshold value, and then starts decreasing. What could be the possible reason for this decrease

Respuesta :

Otras preguntas

27 g of Al will react with how much mass of O2 to produce Al2O3

In the South Pacific Marines used coded messages sent in the _____ language.

Which is a fused sentence? I love drawing because it exercises my imagination. Our pet poodle loves to sit on our laps and take long naps. Pablo took a music cl

the colonists had difficulty coming together to revolt because they: A. were not complety unified. B. had not read the prowar pamphlets. C. had no real reason t

If you were asked to write a summary of the 1968 presidential election, on whom would you likely focus significant attention? Gerald Ford John F. Kennedy Martin

Solve the area formula for a triangle, A = 1/2bh , for h. a) [tex]h= \frac{2b}{A} [/tex] b) [tex]h= \frac{2A}{b} [/tex] c) [tex]h= \frac{A}{2b} [/tex] d) [tex

When Ryan’s alarm clock woke him up at eight in the morning, he hardly suspected that he would never hear it ring again. Which narrative device is used in this

Which type of law governs the relationship between private individuals or companies?

Why would this African ceremonial mask be worn? To cover the face of a criminal To learn from the teachings of a griot To communicate with ancestors To frighten

what makes farming in mountainous areas so difficult without terracing the poor sound quality of mountain land? the rugged landscape of mountain and the lack of

stanfordgoddy stanfordgoddy · Answer 1 · 2020-11-07T21:21:05+01:00

Hi, you've asked an incomplete question. Here's the remaining part of the questions:

What could be the possible reason for this decrease?

A. Even if number of kernels increase, only few of them are used for prediction

B. As the number of kernels increase, the predictive power of neural network decrease

C. As the number of kernels increase, they start to correlate with each other which in turn helps overfitting

D. None of these

Answer:

C. As the number of kernels increase, they start to correlate with each other which in turn helps overfitting

Explanation:

Remember, Kernels refer to computer programs that occupy part of the operating system memory.

By looking closely at the graph we could notice that the positive trend of the test accuracy couldn't be sustained as a result of a constant increase in the width of the neural network.