Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How Google Translate squeezes deep learning onto a phone

googleresearch.blogspot.com

Post reply on HN
Print view View original

41–50 of 99 posts

Re: How Google Translate squeezes deep learning onto a phone

#41

Post by dchichkov »

motoboi wrote: post #15

I am 15 years into this computers thing and this blog post made me feel like "those guys are doing black magic". Neural networks and deep learning are truly awesome technologies.

They are doing matrix multiplications. To pass input a single time through even some very large neural network - it is a relatively fast operation (if compared to training such a network, that is). Training requires data centers and arrays of GPUs. Passing the input through the network - usually you can get away with a single core and vectorized operations. Unless you are doing high resolution computer vision in real time... You can still get away with the single core even then, but that requires some very smart sublinear processing.

Re: How Google Translate squeezes deep learning onto a phone

#42

Post by mattmanser »

Uhhrrr wrote: post #39

I don't get it. They say they use a dictionary, and they say it works without an Internet connection. How can both things be true? I'm pretty sure there's not, say, a Quechua dictionary on my phone.

It doesn't come with all the languages, you have to download them.

Re: How Google Translate squeezes deep learning onto a phone

#43

Post by bduerst »

mattmanser wrote: post #42
Uhhrrr wrote: post #39

I don't get it. They say they use a dictionary, and they say it works without an Internet connection. How can both things be true? I'm pretty sure there's not, say, a Quechua dictionary on my phone.

It doesn't come with all the languages, you have to download them.

I think it's Android only that you can download the language packs, FYI. The language packs + offline maps are super helpful when travelling abroad.

Re: How Google Translate squeezes deep learning onto a phone

#44

Post by rndn »

mattmanser wrote: post #42
Uhhrrr wrote: post #39

I don't get it. They say they use a dictionary, and they say it works without an Internet connection. How can both things be true? I'm pretty sure there's not, say, a Quechua dictionary on my phone.

It doesn't come with all the languages, you have to download them.

You can download the language packs on Android and iOS, and each one is about 4 MB in size.

Re: How Google Translate squeezes deep learning onto a phone

#47

Post by ori_b »

Uhhrrr wrote: post #39

I don't get it. They say they use a dictionary, and they say it works without an Internet connection. How can both things be true? I'm pretty sure there's not, say, a Quechua dictionary on my phone.

Are you sure?

On my desktop, the english dictionary is ~1 megabyte uncompressed, and compresses to ~250k with gzip. The download for Google Translate is somewhere around 30 megabytes.

Re: How Google Translate squeezes deep learning onto a phone

#48

Post by josu »

Uhhrrr wrote: post #39

I don't get it. They say they use a dictionary, and they say it works without an Internet connection. How can both things be true? I'm pretty sure there's not, say, a Quechua dictionary on my phone.

You have to download them beforehand, and offline translating is limited to just a few languages.

Re: How Google Translate squeezes deep learning onto a phone

#50

Post by sp332 »

anantzoid wrote: post #32

Just waiting for the paper to come out that'll detail all the transformations that were done on the training data specifically for the phone and how did they arrive at deciding to use them. > To achieve real-time, we also heavily optimized and hand-tuned the math operations. That meant using the mobile processor’s SIMD instructions and tuning things like matrix multiplies to fit processing into all levels of cache me…

Not fitting into cache just means it will run slower. Why would it crash?
Post reply on HN