Automatic Differentiation.

This is a video that covers Automatic Differentiation.
Attribution-NonCommercial-ShareAlike CC BY-NC-SA
Authors: Matthew Yedlin, Mohammad Jafari
Department of Computer and Electrical Engineering, University of British Columbia.

[@lllepricon7715]

Thank you! It was very informative. I finally understand why we need the Dual Numbers.

[@mattyedlin7292]

Thank you! I usually give a quiz on how to do the dual numbers for the 1 dimensional chain rule for f(g(x)) using dual numbers. Once you have that and the extension to multiple dimensions, we have the kernel for AUTODIFF in the general setting.

[@Tibug]

I always get chills when people hold their whiteboard pen orthogonally.

[@YazlmaGonulVer]

Thanks for the great explanation.

[@AJ-et3vf]

Great video! Thank you!

[@RealMcDudu]

The challenge at 07:00 - the trick is to use the full (here unproven) rule f(x+y\epsilon) = f(x)+y\epsilon df/dx. That way you can also prove the chain rule.

[@splash2981]

Thanks. Anyone know where is the next video for a more complex example?

[@sapito169]

thanks i fanaly understand

[@breeeesh]

Cursed imaginary unit j! ;) ty for the video!

[@gabrielmccartney7975]

Can we use dual numbers for integration?

[@mattyedlin7292]

This is a very interesting question. Dual numbers provide a way of automating differentiation for the sequence of back propagation - local maps. I will think about this more in the coming week

[@gabrielmccartney7975]

@@mattyedlin7292 thank you very much I'm looking forward to that!

[@gabrielmccartney7975]

@@mattyedlin7292 What topics about dual numbers and automatic differentiation has not been studied yet?