Ali Rahimi’s “Random Kitchen Sinks” Video

October 21, 2013 in Compressed Sensing by hundalhh | 3 comments

I don’t have any time these days, so I’m just going to say that the link below is the best machine learning video that I have seen this year.

“Random Kitchen Sinks: Replacing Optimization with Randomization in Learning“.

(Thanks to Ali Rahimi for giving the talk and thanks to Carl for sending me the link.)

Edit 10-29-13:

Check out Igor’s and Alex Smola posts

“Random Kitchen Sinks, Fast Food and other randomized kernel evaluations” (Nuit Blanche)

“Compressed Sensing: Random Features for Large-Scale Kernel Machines” (Nuit Blanche)

“The Neal Kernel and Random Kitchen Sinks” (Adventures in Data Land)

 

 

3 comments

  2. hundalhh’s avatar

    hundalhh on October 29, 2013 at 8:27 am

    Igor,

    Thank you so much for the link. I had looked at your post briefly back in May, but then forgot about it. I really need to devote more time to reading up on the best new algorithms.

    Cheers,
    Hein

    1. Igor Carron’s avatar

      Igor Carron on October 29, 2013 at 9:44 am

      Hein,

      That video is indeed very interesting not just on the RKS aspect but also because it festures some hardware implementation. Few machine learning talks have that quality.

      Cheers,

      Igor.

Comments are now closed.