Deep Learning Book Community | CommonLounge

Deep Learning Book

3343 members.

A supporting forum / community for the book by Ian Goodfellow, Yoshua Bengio and Aaron Courville available at deeplearningbook.org

Invite Members

Ask a question or start a poll…

Link

Poll

CommonLounge is a community of learners who learn together. Get started with the featured resources above, ask questions and discuss related stuff with everyone.

NaN.

discussion

Chapter 6: Deep Feedforward Networks

If you're reading Deep Learning Book, feel free to ask questions, discuss the material or share resources.

Below is a video from a book club in San Francisco, USA discussing this chapter. Presented by Timothee Cour.

Read more…(37 words)

http://deeplearningbook.org/contents/mlp.html

Category: Deep Learning Book

Top comments in this discussion

Herli Menezes25w

there is a very good explanation at StackExchange given by Gabrriel Romon Motivating sigmoid output units in neural networks starting with unnormalized log probabilities linear in $z=w^Th+b$ and $\phi(z)$

Read more… (29 words)

Read more (29 words)

equation 6.22 can be rewritten into:

\frac{e^{yz}}{e^{yz}+e^{(1-y)z}}

Read more… (6 words)

Read (6 words)

How to understand the equation 6.31 , and How does it derive from ?

Read more… (14 words)

Read (14 words)

Show 16 more
comments

2.

discussion

CHAPTER 6: Confused by Algorithm 6.4

I am confused by this line:

$\nabla_{W^{(k)}}J = gh^{(k-1)T}+\lambda\nabla_{W^{(k)}}{\Omega(\theta)}$

This equation is derived by the equation below:

$a^{(k)}= b^{(k)}+W^{(k)}h^{(k-1)}$

But the gradient of $W^{(k)}h^{(k-1)}$ with respect to $W^{(k)}$ should be $h^{(k-1)}$, right?

Why is there a transpose $T$ here?

A matrix calculus problem in backpropagation encount...

Read more…(50 words)

A matrix calculus problem in backpropagation encountered when studying Deep Learning

https://math.stackexchange.com/questions/2970202/a-matrix-calculus-problem-in-backpropagation-encountered-when-studying-deep-lear

Category: Deep Learning Book

3.

discussion

Chapter 10: Sequence Modeling - Recurrent and Recursive Nets

If you're reading Deep Learning Book, feel free to ask questions, discuss the material or share resources.

Below is a video from a book club in San Francisco, USA discussing this chapter. Presented by Ian Goodfellow.

Read more…(37 words)

http://deeplearningbook.org/contents/rnn.html

Category: Deep Learning Book

4.

discussion

Chapter 8: Optimization for Training Deep Models

If you're reading Deep Learning Book, feel free to ask questions, discuss the material or share resources.

Read more…(18 words)

http://deeplearningbook.org/contents/optimization.html

Category: Deep Learning Book

NaN.

discussion

Chapter 7: Regularization for Deep Learning

If you're reading Deep Learning Book, feel free to ask questions, discuss the material or share resources.

Read more…(18 words)

http://deeplearningbook.org/contents/regularization.html

Category: Deep Learning Book

Top comment in this discussion

Michael Bilow2y

In Chapter 7, Page 234, it says:

The new matrix is the same as the original one, but ...

Read more… (123 words)

Read (123 words)

NaN.

discussion

Chapter 9: Convolutional Networks

If you're reading Deep Learning Book, feel free to ask questions, discuss the material or share resources.

Read more…(18 words)

http://deeplearningbook.org/contents/convnets.html

Category: Deep Learning Book

Top comments in this discussion

I don't think so - the indices taken alone can be arbitrary (i.e. he could have written a,b,c or x,y,z for the indices of V). However when we are taking sums we need to ensure that the indices match.

In other words, I can say that V_x,y,zrepresent the ...

Read more… (186 words)

Read more (186 words)

Could someone explain what the last sentence on page 342 means:

"In linear algebra notation, we index into arrays using 1 for the first entry"

I'm rather confused, because say if I wanted to find...

Read more… (103 words)

Read more (103 words)

Bastian Lampe1y

This is on page 342. In the first sentenced, j is referred to as a channel in the input. In the second sentence, j is refe...

Read more… (38 words)

Read more (38 words)

Show 2 more
comments

Load More