Git Product home page Git Product logo

arunmallya.github.io's People

Contributors

arunmallya avatar dependabot[bot] avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

arunmallya.github.io's Issues

Typo error for blog.

Hi arunmallya:

Recently I found your personal website and I learned a lot form your blog, thanks!

I found a typo error in your post "A Quick Introduction to Backpropagation". In the section
"Softmax Layer", the back propagation seems to be this, you missed an item.

m

Your final result is all right, so I think it is just a type error.

Website not displayed correctly

First of all: thank you for your writeups on Backprop in general and the LSTM backprop. It has been a huge help!
I have a problem with your post on LSTM-backprop not being displayed correctly. Right now I am on a Windows7 computer with Firefox, but it has the same problem on different browsers in Linux (Chrome, Firefox):

gitio

As you can see, the most interesting part - the equations - are always cut of. One can zoom out to 50%-60% but then the equations get quite small.

Backward Pass: Input and Gate Computation - II

I'm not sure how the derivative of h^t-1 could be computed, since many equations in z^t relies on h^t-1. Do we compute each function's derivative and add them together to form the derivative of h^t-1?

c^t hat is a^t

image

I'm new to LSTMs, and I'm trying to derive the backpropagation of LSTMs. Your site was very helpful to me.

Thanks

Question about LSTM bp

qq 20161129210749

Could you explain the meaning of += in detail?

You said that

+= above is so that this gradient is added to gradient from time step (t+1) (calculated on next slide)

But there is no gradient from time step t+1 calculated on the next slide...
Thank you very much.

LSTM Backward Pass Question

qq 20170415144527

I don't understand the equation in the red box.


qq 20170415144610
qq 20170415144621
In my opinion, f^t is also indirectly depend on c^(t-1), since f^t is depend on h^(t-1), and h^(t-1) is depend on c^(t-1). What's more, i^t and a^t are also indirectly depend on c^(t-1) for the same reason.


qq 20170415145303
So I think the above equation may not be correct?

Am I right?

Thank you very much!

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.