@danijarh Whoah, congrats Danijar, totally deserved, youโre still the most impressive researcher, collaborator and mentor that I ever had the pleasure of working with! ๐ซถ
Excited to introduce Dreamer 4, an agent that learns to solve complex control tasks entirely inside of its scalable world model! ๐๐ค
Dreamer 4 pushes the frontier of world model accuracy, speed, and learning complex tasks from offline datasets.
co-led with @wilson1yan
Awesome to see model based RL to get the recognition ๐ Predicting text is not everything, AI needs to learn how the actual world works!
It's been an honour to take part in the project, and I can't wait to see where Dreamer goes from here!
Excited to share that DreamerV3 has been published in Nature!
Dreamer solves control tasks by imagining the future outcomes of its actions inside of a continuously learned world model ๐
It's the first agent to find diamonds in Minecraft from scratch without human data! ๐
๐
๐ The first realtime AI that pilots your desktop is here: Ace
Excited to share what I've been helping build at @GeneralAgentsCo over the past few months
Read on โฌ๏ธ
Today I'm launching my new company @GeneralAgentsCo and our first product.
Introducing Ace: The First Realtime Computer Autopilot
Ace is not a chatbot. Ace performs tasks for you.
On your computer. Using your mouse and keyboard.
At superhuman speeds!
Today I'm launching my new company @GeneralAgentsCo and our first product.
Introducing Ace: The First Realtime Computer Autopilot
Ace is not a chatbot. Ace performs tasks for you.
On your computer. Using your mouse and keyboard.
At superhuman speeds!
๐ Excited to share a major update of the DreamerV3 agent!
A couple of smaller changes, more benchmarks, and substantially improved performance.
๐ Main differences from our earlier preprint:
Gather-Attend-Scatter (GATS), a novel module that combines pretrained foundation models operating at different rates into larger multimodal networks.
Paper: https://t.co/BMsWtuZTQ0
I have started working on a new guide, called
The Art of Debugging
which is a brain dump based on almost 3 decades of developing software. So far I have the initial draft of the first chapter:
Fast Debugging Methodology
https://t.co/h8ROAZjqHh
A lot more to come...
I hope you will find it useful and thank you for reading!
@realGeorgeHotz@ESYudkowsky I sincerely hope you have a better one. Like, expand on why your tinybox pal will basically be on your side, and defense stronger than offense, to make a somewhat stable system.
We just put out a statement:
โMitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.โ
Signatories include Hinton, Bengio, Altman, Hassabis, Song, etc.
https://t.co/N9f6hs4bpa
๐งต (1/6)
Today we're releasing Video Prediction Rewards (VIPER ๐), a simple yet powerful method for extracting rewards from video prediction models!
VIPER learns reward functions from raw videos, and generalizes to entirely new domains for which no training data is available
๐งต thread
This document leaked from Google has been gaining attention. Unfortunately it's wrong and right in major ways that should make us seriously reflect on what we're creating
Yes open source models are a huge deal but more open sourcing is NOT the solution
https://t.co/QbYLorDKcx
Introducing DreamerV3: the first general algorithm to collect diamonds in Minecraft from scratch - solving an important challenge in AI. ๐
It learns to master many domains without tuning, making reinforcement learning broadly applicable.
Find out more: https://t.co/7GP0R93Yvc
Excited to announce DreamerV3 ๐, a scalable and general RL algorithm that masters a wide range of applications with fixed hyperparameters!
Applied out of the box, it solves the Minecraft Diamond challenge without human data. ๐
๐ Thread
https://t.co/zTAy1HZvIW
Introducing DreamerV3: the first general algorithm to collect diamonds in Minecraft from scratch - solving an important challenge in AI. ๐
It learns to master many domains without tuning, making reinforcement learning broadly applicable.
Find out more: https://t.co/7GP0R93Yvc