Excited to share LARF
Our large causal transformer with a small flow-matching head predicts next observations and actions, accelerating policy optimization in the learned world model.
This way robots can learn from trillions of trials instead of relying on costly human teleop.๐
At Tau, we develop world-model based reinforcement learning algorithms to learn reliable policies and improve beyond human performance.
Blog about our first world model: https://t.co/4lOXTXYcyU
Big news to share: @a16z is expanding to the UK ๐ฌ๐ง
We plan to open our first international office in London later this year, and will host the next Crypto Startup School there in 2024.
Why the UK? ๐
https://t.co/PQ7GuNEn77
Big news to share: @a16z is expanding to the UK ๐ฌ๐ง
We plan to open our first international office in London later this year, and will host the next Crypto Startup School there in 2024.
Why the UK? ๐
https://t.co/PQ7GuNEn77
@storming0x Thanks for sharing this, I must have spoken to at least a few dozen people who experienced a war room first hand at this point, the level of stress and pressure it has on amazing people is underrepresented in the conversations around security. Thanks for sharing your experience!