正在加载视频...

视频加载失败

For small step sizes, mirror descent is equivalent to a gradient descent on a Hessian manifold.

61,868 次观看 • 2 年前 •via X (Twitter)

7 条评论

LowρUF👁️ 🇵🇦🇬🇱🇨🇦🇲🇽 的头像
LowρUF👁️ 🇵🇦🇬🇱🇨🇦🇲🇽2 年前

If I understood 65% of this instead of 35% of this, I think it'd save ~90hr of my computer running at full tilt in my current research

Samuel Hurault 的头像
Samuel Hurault2 年前

Does mirror descent correspond to a discretization of this flow ?

Gabriel Peyré 的头像
Gabriel Peyré2 年前

Yes, a very smart one actually (for instance it preserves all the thm of speed of convergence as usual gradient descent)

Alireza Mansouri 的头像
Alireza Mansouri2 年前

How do you make this kind of animation? I mean by which software?

AImaster 的头像
AImaster2 年前

fucking cool! better than any pussies I have got.

AImaster 的头像
AImaster2 年前

@readwise save thread @SaveToNotion #thread @threadreaderapp unroll @memdotai mem it

Thread Reader App 的头像
Thread Reader App2 年前

@gabrielpeyre @readwise @SaveToNotion @memdotai @CeoSpaceY Hi! please find the unroll here: Talk to you soon. 🤖

相关视频