https://t.co/5KdePOhYYt
Nikki Haley just walked by us in the West Wing, when we asked if she\u2019d resigned she smiled and walked into the Oval Office.
— Andrew Feinberg (@AndrewFeinberg) October 9, 2018
#BREAKING
Nikki Haley has reportedly resigned (UNCONFIRMED):
“...according to two sources briefed on their conversation. The timing of her departure is still unclear.”
Scoop: Trump has accepted Nikki Haley's resignation https://t.co/vl5W7CNnEi
— Jonathan Swan (@jonathanvswan) October 9, 2018
https://t.co/5KdePOhYYt
Not quite a confirmation from @presssec at this point: \u201cPresident Trump and Ambassador Nikki Haley will meet in the Oval Office at 10:30am this morning. This event will be open to the pool.\u201d
— Jennifer Epstein (@jeneps) October 9, 2018
It’s difficult to imagine Trump has actual positive regard for Haley, who has not been as deferential as he likes.
From Haley’s side, the timing of this announcement seems like a dig at Trump, given midterms are coming and she’s not leaving ASAP.
https://t.co/jFEqNP1Ogb
Whatever is said by Trump and Haley...don't believe it. They are both experts at trying to put a good face on things and then relentlessly pursuing their own agendas.
— David Rothkopf (@djrothkopf) October 9, 2018
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)
