Profile

Cover photo
Muthu Kannan (Manki)
1,713 followers|4,297,245 views
AboutPostsCollectionsPhotosYouTube+1'sReviews

Stream

Muthu Kannan (Manki)

Shared publicly  - 
5
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
 
I've seen a lot of uncritical comments in the tech press lately about how Tesla's massive number of miles driven give them a huge data collection advantage over anyone else when it comes to training self-driving models.

I'd like to see more analysis of this claim.

From http://www.kurzweilai.net/googles-self-driving-car-gathers-nearly-1-gbsec , the Google self-driving car in 2013 gathered 750 MB/s (note: bytes, not bits) of sensor data. It's not clear to me whether this is before- or after-compression of compressible data types (I assume any "image-like" data is probably compressible).

Let's assume this is before-compression. Let's also assume that the overall data stream can be compressed 15:1 without being degraded enough to harm learning from it. (This number is from the "high quality" line on https://en.wikipedia.org/wiki/JPEG#Effects_of_JPEG_compression . JPEG is rather archaic as image compression goes, but then again, not all the data above is image data. This seems like at least a plausible value to do napkin math.) This gives us 50 MB/s of data. That's ~1.5 TB data in an 8 hr. driving day.

Google can easily store this much data onboard and upload it when the cars return to home base. So they can actually use it all for training.

Tesla doesn't have this kind of a "home base" that all customer cars return to, and it's clear they're not storing or uploading anything like 50 MB/s of data. They're not uploading 5 MB/s, or 0.5 MB/s, across their customer base. Neither the network traffic to send this nor the data centers to store it are apparent. So most data Tesla is collecting must be discarded locally, and only a small amount is being sent upstream.

How useful is that small amount? It's hard to say, but given https://twitter.com/elonmusk/status/811738008969842689 , a neural net is involved in at least some areas. From what I know about neural nets, you can distribute a trained model to a local device that can use it with relatively low CPU/memory requirements, but actually training a new model needs to occur in a resource-intensive environment (read: Tesla HQ, not in your car). So the neural nets in question here aren't being trained on the actual car sensor data from all those cars Tesla has sold. It seems unlikely they are being trained on the fragments of data sent back, either, because it's just not going to provide sufficiently high-band input to be useful (you can't train a vision algorithm on a written summary of a movie plot). I'd imagine Tesla is training them on data more like Google's: high-bandwidth data collected from a test fleet.

Note how Musk says "need to get a lot of road time". If all their existing data from their customers' cars was so useful, why couldn't they use that data? The obvious answer is that it's not, in fact, that useful. Either the sensor model is different, or the field data is far too highly compressed to train or validate things. I suspect the latter.

My conclusion is that Tesla's fleet of customer cars likely provides them far less benefit than everyone is uncritically assuming, when it comes to improving their software.

There are a lot of assumptions and leaps in the above. This isn't proof Tesla doesn't have a big advantage. But it does suggest that before the public and press trumpet Tesla's "clear lead" and repeat endlessly how Tesla's millions of miles driven are more than anyone else's, some reporters should dig into the links of the above chain to verify. Has anyone reported on Tesla's data collection bandwidth at various stages in the pipeline, how they compress or sample data, what they transmit back to HQ, how the data is used for model training, etc.? If yes, please link me, as that would be fascinating. If no, then next time you read someone state this data collection advantage as fact, ask yourself how certain that is.

In this as with much of life, more skepticism and critical thinking would be useful.
(Credit: Google) Google's self-driving car gathers 750 megabytes of sensor data per SECOND! That is just mind-boggling to me. Here is a picture of what the
1 comment on original post
1
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
You know your life has changed when a nursery rhyme is stuck in your head.

♪ ♫ bake me a cake as fast as you can ♫ ♪
2
Chris Veerabadran's profile photoMuthu Kannan (Manki)'s profile photoSuresh V Selvaraj's profile photo
9 comments
 
Well, debugging takes time and experience. lol. But cool that you've started. I guess baking that cake is not impossible then! Don't forget to post pics.. whether u make a disaster movie, a comedy movie or a social drama, we would like to see! :D
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
Cool work, scientists. Now please invent an apparatus to refold washed clothes. Thanks! 😄
1
Muthu Kannan (Manki)'s profile photoSuresh V Selvaraj's profile photo
5 comments
 
lol. exactly! Laziness is the mother of invention!
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
2017 starts with a healthy breakfast. Sauteed vegetables with fried eggs.
2
Chris Veerabadran's profile photoMuthu Kannan (Manki)'s profile photo
2 comments
 
Happy new year to you too, +Chris Veerabadran​ 😃
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
Grilled chicken, asparagus, and sweet potato
3
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
This would come handy.
00A7 section. ¶, 00B6 paragraph. ·, 00B7 interpunct, middle dot. ¼, 00BC one-quarter. ½, 00BD one-half. ¾, 00BE three-quarters. ‒, 2012 figure dash. –, 2013 en dash. —, 2014 em dash. ―, 2015 quotation dash. ' 2018 left single quote. ' 2019 right single quote, apostrophe.
2
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
Instead of saying "if you have automatic payments enabled", you could be helpful and say "we'll charge your credit card ending x-1232 on 14th Jan" or "you have not enabled automatic payments". #thatShouldntBeTooHard
1
Add a comment...

Muthu Kannan (Manki)

Shared publicly  - 
 
“All the 10,000 residents decided to carry on cashless transactions, with the installation of 40 card-swiping machines in their village. As Mid-Day reports, everybody, right from the doctor to the barber, will use only plastic money.

It was the initiative of Ranjeet Savarkar, president of the Savarkar Smarak organisation, an NGO helping the village go cashless. With the help of Bank of Baroda, and the Jan Dhan Yojana, villagers will now have the Rupay ATM card.”
4
Add a comment...
Muthu's Collections
Story
Tagline
சர்வம் கிருஷ்ண லீலா
Introduction
Basic Information
Gender
Male
Other names
Manki, குசேலன்
Muthu Kannan (Manki)'s +1's are the things they like, agree with, or want to recommend.
How to View Your Chromebook’s Hardware Specifications and System Informa...
www.howtogeek.com

Google doesn’t provide an easy way to see the storage, RAM, CPU, and other specifications of your Chromebook. But it’s possible to dig all t

நெஞ்சுக்குள்ளே
tamilpaadalvarikal.blogspot.com

திரைப்படம் : கடல் 2013 இயக்குனர் : மணி ரத்னம் நடிப்பு : கவுதம் கார்த்திக், துளசி நாயர் இசையமைப்பு : ஏ. ஆர். ரகுமான் பாடலாசிரியர் : வைரமுத்து

How to Remove a Password from a PDF File in Linux
www.howtogeek.com

You’ve protected a PDF file containing sensitive information with a long, secure password so only the intended party can open it. However, y

Google App v5.10 Adds Option To Open Links In Chrome Custom Tabs
www.androidpolice.com

The Google App's version 5.10 update rolled out to the beta channel yesterday with a new waiting animation, but one change that was lurking

Creating a timelapse clip with avconv | Techedemic
techedemic.com

-y forces avconv to overwrite any file -r 10 creates a clip with 10 frames per seconds (for some versions of avconv you have to specify it t

Night Time Photography Tips
www.picturecorrect.com

Many digital photography beginners I know stick rigidly to daytime shots, apprehensive about their rate of success at night. If you’ve delay

TechSpot
plus.google.com

Technology News, Reviews and Analysis

Cell Size and Scale
learn.genetics.utah.edu

Meter, Centimeter, Millimeter, Micrometer, Nanometer, Angstrom, Picometer. 100 m, 10-2 m, 10-3 m, 10-6 m, 10-9 m, 10-10 m, 10-12 m. 1 m, 0.0

Comcast to bring two-gigabit Internet service to Bay Area, boost existin...
www.pcworld.com

Silicon Valley residents won't have to wait for Google Fiber: The cable giant says it will roll out the service in May.

சித்திரமும் கைப்பழக்கம்
studied.wordpress.com

சித்திரமும் கைப்பழக்கம் செந்தமிழும் நாப்பழக்கம் வைத்ததொரு கல்வி மனப்பழக்கம் – நித்தம் நடையும் நடைப்பழக்கம் நட்பும் தயையும் கொடையும் பிறவிக்

Getting Over jQuery
ponyfoo.com

We've looked at doing some of the things that you can do in native code. So far, we've covered AJAX, event handling, event delegation, DOM q

Replacing jQuery with Vanilla JavaScript - Modern Web
modernweb.com

By Zachary Brady It seems as if jQuery has become synonymous with JavaScript. Since its introduction in 2006, jQuery has caused a revolution

ACM Classic: Reflections on Trusting Trust
cm.bell-labs.com

represents a string with the character "\n," representing the new line character. [figure 2] FIGURE 2. Figure 2 is an idealization of the co

ஆறாம் திணை - 01
aaraamthinay.blogspot.com

கு.சிவராமன் அடிப்படையில் சித்த மருத்துவர். உடல் நலனுக்கும் மன நலனுக்கும் சுற்றுச்சூழலும் இயற்கையோடு இயைந்த வாழ்வும் எந்த அளவுக்கு முக்கியம்

DOGHOUSE | Two Thousand And Fifteen
thedoghousediaries.com

Hey people! Forekast just launched! It's a project we've been working on. If you wanna know more, here's our blog post. << · >> << · >> Tw

Vaishali has an exotic menu for its location—they make foods like naan stuffed with fish/chicken/prawn. We tried parotta stuffed with fish and prawn. Both were delicious and the price was pretty cheap. Their regular parotta was also good. Great service. They provide free bottled water as well. Highly recommended.
Public - in the last week
reviewed in the last week
You can get pretty much anything including broccoli, pasta, and sometimes avocado. Fresh vegetables and fruits are available throughout the year. Built many decades ago, so the roads are narrow and it can get congested at peak hours. Don't take your car to this market; use motorcycles or public transport instead.
Public - a week ago
reviewed a week ago
Consistently good food. (Except for briyanis on weekdays. Their weekend briyanis are so much better.) Friendly people. One of the best places for South Indian food in South Bay.
Public - a month ago
reviewed a month ago
We stayed in the queen studio cabin. It was clean and cozy. The kitchen had a lot of utensils to cook with, which was nice. Managers of the property Bob and Laura were friendly and helpful. The bathroom was a bit too small; especially the shower was a tight fit. Overall, or stay here was pleasant.
Public - a month ago
reviewed a month ago
48 reviews
Map
Map
Map
Public - 3 weeks ago
reviewed 3 weeks ago
Good place for shopping South Indian grocery. They even sell briyanis and puffs on weekends. Staff are friendly people as well.
Public - a month ago
reviewed a month ago
Rooms were fine; included cold breakfast was okay. None of this was great, but when looked at the context of the price (I paid a little over $60 for a night) this is great value for money.
Public - 3 months ago
reviewed 3 months ago