Simply untrue. They will still be teaching remotely, which is often more difficult than in person
If they can't do that because of lack of tech that's partly the govt's fault and it should have been better prepared by now
Part of the problem with the debate around schools is the obscure nature of the accusations against teachers
What exactly are teachers being accused of?
...
Simply untrue. They will still be teaching remotely, which is often more difficult than in person
If they can't do that because of lack of tech that's partly the govt's fault and it should have been better prepared by now
AFAIK virus risk in schools broadly matches the wider population (happy to be corrected), which is rising rapidly
If there's increased risk everywhere, there is in schools too
In some schools it is banned, in others simply frowned upon
Add to that almost 9 million pupils: https://t.co/DFV8X3Z7Cv
That's a lot of people (~18% of the population in England) mixing closely, even with mitigations in place, and should concern us all
This is a malicious accusation against people who dedicate their lives to not just teaching but providing a whole range of pastoral care to children. I give it no value.
Personally, I don't think so
If there were sufficient testing and mandatory use of PPE until widespread vaccination, the level of risk might be different (again govt shld shoulder blame)
Some may be comfortable working in schools, that's a personal choice
But the right to be safe and secure at work is basic and not too much to ask, not even (in fact, especially) in a pandemic
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)
