What if you want to startup, but you have no technical skills like web development, coding, etc?
😧
More from All
How can we use language supervision to learn better visual representations for robotics?
Introducing Voltron: Language-Driven Representation Learning for Robotics!
Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z
🧵👇(1 / 12)
Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.
Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)
The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).
The secret is *balance* (3/12)
Starting with a masked autoencoder over frames from these video clips, make a choice:
1) Condition on language and improve our ability to reconstruct the scene.
2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)
By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.
Why is the ability to shape this balance important? (5/12)
Introducing Voltron: Language-Driven Representation Learning for Robotics!
Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z
🧵👇(1 / 12)
Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.
Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)
The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).
The secret is *balance* (3/12)
Starting with a masked autoencoder over frames from these video clips, make a choice:
1) Condition on language and improve our ability to reconstruct the scene.
2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)
By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.
Why is the ability to shape this balance important? (5/12)
You May Also Like
Took me 5 years to get the best Chartink scanners for Stock Market, but you’ll get it in 5 mminutes here ⏰
Do Share the above tweet 👆
These are going to be very simple yet effective pure price action based scanners, no fancy indicators nothing - hope you liked it.
https://t.co/JU0MJIbpRV
52 Week High
One of the classic scanners very you will get strong stocks to Bet on.
https://t.co/V69th0jwBr
Hourly Breakout
This scanner will give you short term bet breakouts like hourly or 2Hr breakout
Volume shocker
Volume spurt in a stock with massive X times
Do Share the above tweet 👆
These are going to be very simple yet effective pure price action based scanners, no fancy indicators nothing - hope you liked it.
https://t.co/JU0MJIbpRV
52 Week High
One of the classic scanners very you will get strong stocks to Bet on.
https://t.co/V69th0jwBr
Hourly Breakout
This scanner will give you short term bet breakouts like hourly or 2Hr breakout
Volume shocker
Volume spurt in a stock with massive X times
These 10 threads will teach you more than reading 100 books
Five billionaires share their top lessons on startups, life and entrepreneurship (1/10)
10 competitive advantages that will trump talent (2/10)
Some harsh truths you probably don’t want to hear (3/10)
10 significant lies you’re told about the world (4/10)
Five billionaires share their top lessons on startups, life and entrepreneurship (1/10)
I interviewed 5 billionaires this week
— GREG ISENBERG (@gregisenberg) January 23, 2021
I asked them to share their lessons learned on startups, life and entrepreneurship:
Here's what they told me:
10 competitive advantages that will trump talent (2/10)
To outperform, you need serious competitive advantages.
— Sahil Bloom (@SahilBloom) March 20, 2021
But contrary to what you have been told, most of them don't require talent.
10 competitive advantages that you can start developing today:
Some harsh truths you probably don’t want to hear (3/10)
I\u2019ve gotten a lot of bad advice in my career and I see even more of it here on Twitter.
— Nick Huber (@sweatystartup) January 3, 2021
Time for a stiff drink and some truth you probably dont want to hear.
\U0001f447\U0001f447
10 significant lies you’re told about the world (4/10)
THREAD: 10 significant lies you're told about the world.
— Julian Shapiro (@Julian) January 9, 2021
On startups, writing, and your career: