Dr. Fei-Fei Li

And in that neural network, it has all the parameters, the mathematical parameters that's already learned now.

And in that neural network, it has all the parameters, the mathematical parameters that's already learned now.

You can take this, and now you have a new sentence come in.

You can take this, and now you have a new sentence come in.

And then it goes through this model.

And then it goes through this model.

Because it has all the parameter it has learned, it predicts what I should say given the new sentence, like, hello, Hala, how is your breakfast today?

Because it has all the parameter it has learned, it predicts what I should say given the new sentence, like, hello, Hala, how is your breakfast today?

And it would predict I had a great breakfast today or whatever.

And it would predict I had a great breakfast today or whatever.

So that's how it's going to be used.

So that's how it's going to be used.

It's because math, the way we do math in human mind is different from the way we do language.

It's because math, the way we do math in human mind is different from the way we do language.

Language has a very clear pattern of sequence to sequence.

Language has a very clear pattern of sequence to sequence.

Like I say the word how, you know, the word are and you typically follow, but sometimes it doesn't, right?

Like I say the word how, you know, the word are and you typically follow, but sometimes it doesn't, right?

So I have to learn these patterns.

So I have to learn these patterns.