Convexity and NN
Lectures · 72 · 2 · 12 Jan '21
Hello, I was wondering why we don't use a convex activation (whose compositions would also be co…
Representation power
Lectures · 142 · 3 · 8 Jan '21
Dear TAs, I'm struggling to understand why on a bounded domain, neural nets can **not** approxim…
Logistic regression equation on the first page
Lectures · 86 · 5 · 29 Dec '20
Hello, I wanted to ask if the equation in the photo is correct to have y in the power in the den…
Complexity
Lectures · 113 · 2 · 24 Dec '20
As seen during the lecture, the complexity of the forward pass is \(O\left(LK^{2}\right)\). In t…
weight sharing
Lectures · 84 · 1 · 22 Dec '20
Dear TAs, ![Screenshot from 2020-12-22 18-07-36..jpg](https://oknoname-crm1.s3.amazonaws.com/spi…
NN how to choose architecture
Lectures · 147 · 2 · 26 Nov '20
Hello, I have a question regarding the neural net (or convolutional neural nets) architecture. T…
Transition in the backward pass
Lectures · 113 · 7 · 17 Nov '20
Hi, I don't fully understand a part of the backward pass justification we are making in the cour…
Pytorch tensor for RELU
Lectures · 81 · 2 · 11 Nov '20
Hello, I try to solve a ML problem using a NN with RELU activation function. my data points loo…
Vanishing gradient
Lectures · 78 · 3 · 10 Nov '20
Hi, I didn't quite understand the problem of having an activation function with vanishing gradie…