๊ฐ€์ž… ํ›„ ์ดˆ๋Œ€ ๋งํฌ๋ฅผ ๊ณต์œ ํ•˜๋ฉด ๋™์˜์ƒ ์žฌ์ƒ ๋ฐ ์ดˆ๋Œ€ ๋ณด์ƒ์„ ๋ฐ›์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

Jeff Clune
@jeffclune
Co-founder, Recursive. Professor, CS, U. British Columbia. CIFAR AI Chair, Vector Institute. | ML, AI, deep RL, deep learning, AI-Generating Algorithms (AI-GAs)
๊ฐ€์ž… July 2010
448 ํŒ”๋กœ์ž‰ ์ค‘    33.5K ํŒฌ
Amazing and rewarding to see our small-team research effort (led by @ashrewards , @jparkerholder, @_rockt et al.) take on a life of its own as an ever-improving flagship Google product! For me personally, a crazy idea I proposed in my 2019 AI-GA paper (a NN as the entire environment including pixels, sound, etc.) barely worked in Genie1 (2024), worked decently in Genie 2 (8 months later), became incredible in Genie3 (9 months later), and now Genie4 seems ready for Hollywood and VR. What an arc to observe in such a short time. Wild times! Now imagine open-ended learning in Genie. โœจ ๐Ÿš€
๋” ๋ณด๊ธฐ
This is Genie 4, our most realistic and interactive world model yet. ๐ŸŒ Explore dynamic environments in real time to see how the scene responds, sounds, and how the people in it move, react, and speak.
๋” ๋ณด๊ธฐ