Re: Google DeepMind News and Discussions
Posted: Sat May 14, 2022 4:09 pm
A community of futurology enthusiasts
https://www.futuretimeline.net/forum/
https://www.futuretimeline.net/forum/viewtopic.php?f=16&t=152
Someone else has invented the term "subhuman AGI" for Gato.funkervogt wrote: ↑Thu May 12, 2022 4:18 pm DeepMind builds an AI that is neither narrow nor general. "Narrowly general" or "multipurpose" might be the right term.
https://storage.googleapis.com/deepmind ... 0Agent.pdfInspired by progress in large-scale language modeling, we apply a similar approach towards building a
single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a
multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights
can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based
on its context whether to output text, joint torques, button presses, or other tokens. In this report we
describe the model and the data, and document the current capabilities of Gato.
Jeez.funkervogt wrote: ↑Mon May 16, 2022 9:16 pmSomeone else has invented the term "subhuman AGI" for Gato.funkervogt wrote: ↑Thu May 12, 2022 4:18 pm DeepMind builds an AI that is neither narrow nor general. "Narrowly general" or "multipurpose" might be the right term.
https://storage.googleapis.com/deepmind ... 0Agent.pdfInspired by progress in large-scale language modeling, we apply a similar approach towards building a
single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a
multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights
can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based
on its context whether to output text, joint torques, button presses, or other tokens. In this report we
describe the model and the data, and document the current capabilities of Gato.
https://www.lesswrong.com/posts/TwfWTLh ... -early-agi
Is it weird that even in the context of AI "subhuman" still sounds so...problematic?funkervogt wrote: ↑Mon May 16, 2022 9:16 pmSomeone else has invented the term "subhuman AGI" for Gato.funkervogt wrote: ↑Thu May 12, 2022 4:18 pm DeepMind builds an AI that is neither narrow nor general. "Narrowly general" or "multipurpose" might be the right term.
https://storage.googleapis.com/deepmind ... 0Agent.pdfInspired by progress in large-scale language modeling, we apply a similar approach towards building a
single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a
multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights
can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based
on its context whether to output text, joint torques, button presses, or other tokens. In this report we
describe the model and the data, and document the current capabilities of Gato.
https://www.lesswrong.com/posts/TwfWTLh ... -early-agi