Bayesian Thinking
Programming Throwdown
00:00
Then, You Know, It's a Logical Regression, Right?
Deep learning can be used to train click probability models, but it's not as deeply a complex with layers of abstraction as is image recognition. The repmatorization tricks are something that most people could do am with really any data without having to fit distributions. But hey, i'm sure google will use it, i to kind of squeeze an extra point, one % efficiency out of there, out of their clicks and yey'll get translated to some huge amount of money.
Transcript
Play full episode
