Sam Altman: CEO of OpenAI calls for the US to regulate artificial intelligence He gave several suggestions for how a new agency in the US could regulate the industry—including “giving out and taking away permits for AI companies” Discussion on r/ML: https://
reddit.com/r/MachineLearn
ing/comments/13jhduh/n_sam_altman_ceo_of_openai_calls_for_us_to/
…
@hardmaru
-
Sam Altman advocates for US AI regulation and industry permits
By
–
-
DeepMind’s Robot Based on Open-Platform Robotis OP3 Humanoid
By
–
DeepMind’s robot is based on the off-the-shelf Robotis OP3, and open-platform humanoid project. https://
emanual.robotis.com/docs/en/platfo
rm/op3/introduction/
… Reference: https://
arxiv.org/abs/2304.13653 -
Hyperbolic Deep Reinforcement Learning ICLR2023 Paper and Blog Post
By
–
Updated #ICLR2023 final version of the Hyperbolic Deep Reinforcement Learning paper: https://
openreview.net/forum?id=TfBHF
Lgv77
… Michael Bronstein (
@mmbronstein
) wrote a blog post about their fantastic idea: https://
towardsdatascience.com/hyperbolic-dee
p-reinforcement-learning-b2de787cf2f7
… -
Real-world MuJoCo humanoid agents in action
By
–
These are literally IRL versions of MuJoCo humanoid agents…
— hardmaru (@hardmaru) 30 avril 2023
Looking forward to see the 2 vs 2 version: https://t.co/Glvcr9SUjPThese are literally IRL versions of MuJoCo humanoid agents… Looking forward to see the 2 vs 2 version:
-
Deep RL Synthesis of Movement Skills for Miniature Humanoid Robot
By
–
This is such a cool project:
— hardmaru (@hardmaru) 29 avril 2023
“We investigate whether Deep RL is able to synthesize sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be composed into complex behavioral strategies in dynamic environments.”
Video of behavior and skills → pic.twitter.com/COIdyFpZNRThis is such a cool project: “We investigate whether Deep RL is able to synthesize sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be composed into complex behavioral strategies in dynamic environments.” of behavior and skills →
-
DeepMind’s Bipedal Robot Soccer Players with Deep Reinforcement Learning
By
–
Pushing around these little robot soccer players, from DeepMind’s “Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning” paper.https://t.co/KzXryyfhCHhttps://t.co/VOnhLk0DyY pic.twitter.com/GDcCuRm8KM
— hardmaru (@hardmaru) 29 avril 2023Pushing around these little robot soccer players, from DeepMind’s “Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning” paper. https://
arxiv.org/abs/2304.13653 https://
sites.google.com/view/op3-soccer -
Try out DeepFloyd IF model on Hugging Face Spaces
By
–
Try out the model here → https://
huggingface.co/spaces/DeepFlo
yd/IF
… -
DeepFloyd IF: Open-source text-to-image model with zero-shot inpainting
By
–
DeepFloyd IF: An open-source text-to-image model by our @DeepfloydAI team @StabilityAI
— hardmaru (@hardmaru) 28 avril 2023
Check out the examples, with amazing zero-shot inpainting results at high resolution: https://t.co/9TbHDzyGOy pic.twitter.com/YjIAyROCpWDeepFloyd IF: An open-source text-to-image model by our @DeepfloydAI team @StabilityAI Check out the examples, with amazing zero-shot inpainting results at high resolution: https://
github.com/deep-floyd/IF -
Open AI Models as Better Research Baselines Than Closed Systems
By
–
Closed AI Models Make Bad Baselines @annargrs “The more people cite and study the best available open solutions, the more we incentivize open and transparent research, and the more likely it is that the next open solution will be much better.”
-
Albert Einstein as James Bond SDXL Image Generation
By
–
“Albert Einstein as James Bond” #SDXL https://
clipdrop.co/stable-diffusi
on
…