In a CNN, what factors most significantly affect the number of parameters in a convolutional layer?
Added by Deborah E.
Step 1
A larger filter size will result in more parameters as there are more weights to learn. Show more…
Show all steps
Your feedback will help us improve your experience
Akash M and 63 other AP CS educators are ready to help you.
Ask a new question
Labs
Want to see this concept in action?
Explore this concept interactively to see how it behaves as you change inputs.
Key Concepts
Recommended Videos
Consider a CNN that takes in 32 Ă— 32 grayscale images and has asingle convolution layer with three 5 Ă— 5 convolution filters (withoutboundary padding).(a) Draw a sketch of the input and first hidden layer similar toFigure 10.8.(b) How many parameters are in this model?(c) Explain how this model can be thought of as an ordinary feedforwardneural network with the individual pixels as inputs, andwith constraints on the weights in the hidden units. What arethe constraints?(d) If there were no constraints, then how many weights would therebe in the ordinary feed-forward neural network in (c)?Please draw the sketch. Original answers only.
Akash M.
What are typical reasons for overfitting? Please list at least three reasons
Adi S.
Consider a CNN that takes in 32 Ă— 32 grayscale images and has a single convolution layer with three 5 Ă— 5 convolution filters (without boundary padding). (a) Draw a sketch of the input and first hidden layer similar to Figure 10.8. (b) How many parameters are in this model? (c) Explain how this model can be thought of as an ordinary feedforward neural network with the individual pixels as inputs, and with constraints on the weights in the hidden units. What are the constraints? (d) If there were no constraints, then how many weights would there be in the ordinary feed-forward neural network in (c)?
Recommended Textbooks
Computer Science and Information Technology
Introduction to Programming Using Python
Computer Science - An Overview
Transcript
Watch the video solution with this free unlock.
EMAIL
PASSWORD