Jascha Sohl-Dickstein's banner
Jascha Sohl-Dickstein's profile picture

Jascha Sohl-Dickstein

@jaschasd • 29,896 subscribers

Member of the technical staff @ Anthropic. Most (in)famous for inventing diffusion models. AI + physics + neuroscience + dynamics.

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.
1:12
jaschasd's profile picture

Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.

Jascha Sohl-Dickstein

1,767,493 Aufrufe • vor 2 Jahren

The boundary between trainable and untrainable neural network hyperparameter configurations is *fractal*! And beautiful! Here is a grid search over a different pair of hyperparameters -- this time learning rate and the mean of the parameter initialization distribution.
1:09
jaschasd's profile picture

The boundary between trainable and untrainable neural network hyperparameter configurations is *fractal*! And beautiful! Here is a grid search over a different pair of hyperparameters -- this time learning rate and the mean of the parameter initialization distribution.

Jascha Sohl-Dickstein

250,458 Aufrufe • vor 2 Jahren

Keine weiteren Inhalte verfügbar