Profile cover photo
Profile photo
srivas venkatesh
srivas's interests
View all
srivas's posts

Post has shared content
Using t-Distributed Stochastic Neighbor Embedding (t-SNE) to visualize high dimensional data is a powerful way to communicate complex information in a more intuitive way. However, while the t-SNE technique is very useful, the visualizations images can be tempting to misread.

In the article below, +Martin Wattenberg and +Fernanda Viegas of the Google Brain team ( and UX Engineer +Ian Johnson walk through a series of simple examples to illustrate what t-SNE diagrams can and cannot show.

Post has attachment
China’s new quantum satellite will try to teleport data outside the bounds of space and time and create an unbreakab…

Post has shared content
Tensor Processing Units (TPUs)
I'm very excited that we can finally discuss this in public. Today at Google I/O +Sundar Pichai revealed the TPU (Tensor Processing Unit), a custom ASIC that Google has designed and built specifically for machine learning applications. We've had TPUs deployed in Google datacenters for more than a year, and they are an order of magnitude faster and more power efficient per operation than other computational solutions for the kinds of models we are deploying to improve our products. This computational speed allows us to use larger, more powerful machine learned models, expressed and seemlessly deployed using TensorFlow ( into our products, and to deliver the excellent results from those models in less time.

TPUs are used on every Google Search to power RankBrain (, they were a key secret ingredient in the recent AlphaGo match against Lee Sedol, they are used for speech and image recognition, and they are powering a growing list of other smart products and features.

+Norm Jouppi and the rest of the team that developed this ASIC did a fabulous job, and it's great to see it discussed in public!

Blog post:

Link to the part of the keynote where Sundar discusses TPUs:

WSJ article:

Edit: Added a link and some text.

Post has shared content
Learn more about the technology behind Allo, a new smart messaging app that uses the power of neural networks and Google Search to make your text conversations easier and more productive.

Post has shared content
From the article: There’s both the raw data from the detectors (so you can verify the results) and also “derived” datasets that are more easy to work with — and don’t worry, CERN is providing the tools to do so, as well. There’s a whole CERN Linux environment ready for booting up in a virtual machine, and a bunch of scripts and apps (some are on GitHub, too).

Post has shared content
Please Keep Your Eyes On The Cross

Staring at the cross causes your peripheral vision to confuse your brain, which then combines features from the different faces and makes them look distorted! XD
Animated Photo

Post has shared content
There's an old bit of elementary-school folklore that you can't fold a piece of paper more than six times. And there's a mathematical theorem that shows exactly how many times you really can fold paper. And then there's the +Hydraulic Press Channel on YouTube, which decided to violate all the laws of mathematics and physics by brute force.

The result is a rather fascinating display of physics at work. I have a theory as to what's going on here, but I would need a hydraulic press, a thermocouple, and an electron microscope to be sure. (Anyone have these sitting around? This would be fun)

You can see the video, the math, and my proposed explanation here:

Thanks to +Chris Colohan for finding this, and posing the challenging question of just what's going on here.
Animated Photo

Post has shared content

Post has shared content
Famous Chunkies credit to Alex solis
edited by me
14 Photos - View album
Wait while more posts are being loaded