What if we set GPT-4 free in Minecraft? ⛏️
I’m excited to announce Voyager, the first lifelong learning agent that plays Minecraft purely in-context. Voyager continuously improves itself by writing, refining, committing, and retrieving *code* from a skill library.
GPT-4 unlocks a new paradigm: “training” is code execution rather than gradient descent. “Trained model” is a codebase of skills that Voyager iteratively composes, rather than matrices of floats. We are pushing no-gradient architecture to its limit.
Voyager rapidly becomes a seasoned explorer. In Minecraft, it obtains 3.3× more unique items, travels 2.3× longer distances, and unlocks key tech tree milestones up to 15.3× faster than prior methods.
We open-source everything. Let generalist agents emerge in Minecraft! Welcome you all to try today: https://t.co/1d3YocozsI
Paper: https://t.co/JcWEasgtyI
Code: https://t.co/KsvVf7rcl0
Deep dive with me: 🧵
This bar chart has attracted the attention of the richest man in the world. Let me walk you through how I would interpret it as a statistician (and a human).
@sonukg4india Point being that binomial prob solution(as others have done) is a good way to think about this problem and that naturally leads to a multinomial solution for the generalized version of this problem.
@sonukg4india This problem can be generalized to n people out of which n1 have bday on specific day, n2 have bday on another specific day, n3 on another specific day. Rest (I.e. n - n1 - n2 - n3) have bdays within the remaining days.
The solution can be expressed using multinomial prob.
True creation eventually becomes independent of its creator. Evolution takes the original product far from its origin, with an uncertain outcome.
– Dune, The Machine Crusade
Why does ChatGPT work so well? Is it “just scaling up GPT-3” under the hood? In this 🧵, let’s discuss the “Instruct” paradigm, its deep technical insights, and a big implication: “prompt engineering” as we know it may likely disappear soon:👇
Unless you operated large systems, it’s hard to appreciate both how they work, and how they break.
Twitter won’t go down with a big b, and it won’t go down irreparable.
Things will work fine for a while - thanks to no new deploys! Then, here’s what’s likely to happen (cont’d):