Barlow Twins: a new super-simple self-supervised method to train joint-embedding architectures (aka Siamese nets) non

Basic idea: maximize the normalized correlation between a variable in the left branch and the same var in the right branch, while making the normalized cross-correlation between one var in the left branch and all other vars in the right branch as close to zero as possible.
2/N
In short: the loss tries to make the normalized cross-correlation between the embedding vectors coming out of the left branch and the right branch as close to the identity matrix as possible.
3/N
The 2 branches are always fed with differently-distorted version of the same image, and there is no need for dissimilar training pairs.

The objective makes the embedding vectors of the two branches as similar as possible, while maximizing their information content.
4/N
No contrastive samples, no huge batch size (optimal is 128), nor predictor, no moving-average weights, no vector quantization, nor cut gradients in one of the branches.
5/N
Competitive results on ImageNet with a linear classifier head.
Great results on semi-supervised ImageNet in the low labeled-data regime and on transfer tasks.
6/N
Results on ImageNet with linear classifier head
7/N
Results with 1% and 10% of ImagNet labeled images
8/N
Results on transfer tasks.
9/N
Arch is standard ResNet50 with 2048-D feature vec.
But contrary to others, the embedding size (projector output) is larger. The perf keeps going up as the embedding dim grows (we stopped at 16384).
Probably cause the feature vars are made independent, not just decorrelated.
10/N
Why Barlow? Horace Barlow was a pioneer of visual neuroscience who proposed the idea that the brain tries to minimize redundancy in representations.

By Jure Zbontar, Li Jing, Ishan Misra, yours truly, and Stéphane Deny.
All from FAIR.
To appear at ICML 2021
11/N
Don't you just hate slicing what would be a decent-size post into threaded thin tweets?
12/N
No, really. Don't you hate reading those long thread slices?
If you do, you could just read my Facebook post:
https://t.co/dQii7BEPQ5
13/N
N=13
Typo: optimal batch size is 1024, not 128.
14/13 (haha).

More from All

APIs in general are so powerful.

Best 5 public APIs you can use to build your next project:

1. Number Verification API

A RESTful JSON API for national and international phone number validation.

🔗
https://t.co/fzBmCMFdIj


2. OpenAI API

ChatGPT is an outstanding tool. Build your own API applications with OpenAI API.

🔗 https://t.co/TVnTciMpML


3. Currency Data API

Currency Data API provides a simple REST API with real-time and historical exchange rates for 168 world currencies

🔗 https://t.co/TRj35IUUec


4. Weather API

Real-Time & historical world weather data API.

Retrieve instant, accurate weather information for
any location in the world in lightweight JSON format.

🔗 https://t.co/DCY8kXqVIK

You May Also Like

One of the most successful stock trader with special focus on cash stocks and who has a very creative mind to look out for opportunities in dark times

Covering one of the most unique set ups: Extended moves & Reversal plays

Time for a 🧵 to learn the above from @iManasArora

What qualifies for an extended move?

30-40% move in just 5-6 days is one example of extended move

How Manas used this info to book


Post that the plight of the


Example 2: Booking profits when the stock is extended from 10WMA

10WMA =


Another hack to identify extended move in a stock:

Too many green days!

Read
The first ever world map was sketched thousands of years ago by Indian saint
“Ramanujacharya” who simply translated the following verse from Mahabharat and gave the world its real face

In Mahabharat,it is described how 'Maharishi Ved Vyasa' gave away his divine vision to Sanjay


Dhritarashtra's charioteer so that he could describe him the events of the upcoming war.

But, even before questions of war could begin, Dhritarashtra asked him to describe how the world looks like from space.

This is how he described the face of the world:

सुदर्शनं प्रवक्ष्यामि द्वीपं तु कुरुनन्दन। परिमण्डलो महाराज द्वीपोऽसौ चक्रसंस्थितः॥
यथा हि पुरुषः पश्येदादर्शे मुखमात्मनः। एवं सुदर्शनद्वीपो दृश्यते चन्द्रमण्डले॥ द्विरंशे पिप्पलस्तत्र द्विरंशे च शशो महान्।

—वेद व्यास, भीष्म पर्व, महाभारत


Meaning:-

हे कुरुनन्दन ! सुदर्शन नामक यह द्वीप चक्र की भाँति गोलाकार स्थित है, जैसे पुरुष दर्पण में अपना मुख देखता है, उसी प्रकार यह द्वीप चन्द्रमण्डल में दिखायी देता है। इसके दो अंशो मे पीपल और दो अंशो मे विशाल शश (खरगोश) दिखायी देता है।


Meaning: "Just like a man sees his face in the mirror, so does the Earth appears in the Universe. In the first part you see leaves of the Peepal Tree, and in the next part you see a Rabbit."

Based on this shloka, Saint Ramanujacharya sketched out the map, but the world laughed
"I really want to break into Product Management"

make products.

"If only someone would tell me how I can get a startup to notice me."

Make Products.

"I guess it's impossible and I'll never break into the industry."

MAKE PRODUCTS.

Courtesy of @edbrisson's wonderful thread on breaking into comics –
https://t.co/TgNblNSCBj – here is why the same applies to Product Management, too.


There is no better way of learning the craft of product, or proving your potential to employers, than just doing it.

You do not need anybody's permission. We don't have diplomas, nor doctorates. We can barely agree on a single standard of what a Product Manager is supposed to do.

But – there is at least one blindingly obvious industry consensus – a Product Manager makes Products.

And they don't need to be kept at the exact right temperature, given endless resource, or carefully protected in order to do this.

They find their own way.