Published on November 15, 2016 by Google

Check out g.co/aiexperiments to learn more.

This experiment helps visualize what’s happening in machine learning. It allows coders to see and explore their high-dimensional data. The goal is to eventually make this an open-source tool within TensorFlow, so that any coder can use these visualization techniques to explore their data. g.co/aiexperiments

Built by Daniel Smilkov, Fernanda Viégas, Martin Wattenberg, and the Big Picture team at Google.

More resources:
www.tensorflow.org

Leave a Reply

39 Comments on "A.I. Experiments: Visualizing High-Dimensional Space"

Notify of
avatar

eddie long
Guest
eddie long
16 days 22 hours ago

this is a great explanation! although i do have a quick question, so if we treat a pixel as a dimension doesnt that limit the usage of this technique to simple images? since a more complex image would have many more pixels and thus the dimensions would be massive, wouldn't this hinder the processing time and cluster formation, aka it would take forever for it to form clusters, just a question, if anyone with more experience with ML can chime that would be awesome!

A nimation
Guest
A nimation
17 days 19 hours ago

How much space does it take up? if it learns that fast then it needs alot of memory to store all that data

Dylan Cope
Guest
Dylan Cope
6 days 13 hours ago

A nimation probably not as much as you think, consider how computer deal with gigabytes worth of data – that's one hundred thousand million bytes, a single dimension won't take up more than a maximum of a couple bytes and even highly dimensional data doesn't tend to beyond thousands.Most of the data is the parameters of the model learned from the data, the easiest way to think about this is by considering the simplest case of splitting one dimensional data by some threshold. That single threshold value is a parameter.

RyonRykal
Guest
RyonRykal
28 days 14 hours ago

Skynet is near!

Gabriel Kwiecinski Antunes
Guest
Gabriel Kwiecinski Antunes
1 month 4 days ago

Não tem como não reconhecer esse sotaque. Kkkkkkkk

ti inu
Guest
ti inu
1 month 9 days ago

Hmm interesting Google and thank you for sharing.

EpicClone5082
Guest
EpicClone5082
1 month 12 days ago

I bet it can't read my bad handwriting

TheAether
Guest
TheAether
1 month 21 days ago

This is so. cool. °v° Thank you for making all of this stuff publicly available!

mastersoftoday
Guest
mastersoftoday
1 month 22 days ago

Hey, can you just hang on with this thing? I translate documents for a living and this is pretty scary. Please stop it, Google!

Fandyus CZ
Guest
Fandyus CZ
1 month 25 days ago

SkyNet is coming soon, pals…

Mikael Murstam
Guest
Mikael Murstam
1 month 26 days ago

Nice explanation. It all makes sense why you built a tensor accelerating unit now.

1 2 3
wpDiscuz