Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

There is now a nice video tutorial that explains how to reproduce our SIGGRAPH 2022 DeepPhase research for learning a quadruped motion controller with the AI4Animation framework. Thanks to Paul Starke! Video: GitHub:

Sebastian Starke

3,285 subscribers

12,451 views • 3 years ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

1 Comments

Leonardo Martins3 years ago

@ps_starke This is an amazing job. Can i use it to train my models? Is there a tutorial on how to change to a biped pipeline?

Related Videos

We are excited to share our #CORL2024 paper (oral) on "Learning Quadruped Locomotion Using Differentiable Simulation" done in collaboration with Sangbae Kim Massachusetts Institute of Technology (MIT). We present a new way to learn to walk in minutes without parallelization, outperforming PPO in sample efficiency! PDF: Video: We present a new framework for learning quadruped locomotion. By leveraging differentiable simulation for policy optimization, our approach achieves fast convergence and stable training, significantly outperforming model-free #ReinforcementLearning methods like PPO in sample efficiency. The key enabler is to combine a high-fidelity, non-differentiable simulator for forward dynamics with a simplified surrogate model for gradient backpropagation. Our framework enables learning quadruped walking in simulation in minutes without parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills on challenging terrains in minutes. This work highlights one of the first successful real-world applications of differentiable simulation for quadruped robots, offering a compelling alternative to traditional RL methods. Kudos to Yunlong Song! UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) Massachusetts Institute of Technology (MIT)MechE

We are excited to share our #CORL2024 paper (oral) on "Learning Quadruped Locomotion Using Differentiable Simulation" done in collaboration with Sangbae Kim Massachusetts Institute of Technology (MIT). We present a new way to learn to walk in minutes without parallelization, outperforming PPO in sample efficiency! PDF: Video: We present a new framework for learning quadruped locomotion. By leveraging differentiable simulation for policy optimization, our approach achieves fast convergence and stable training, significantly outperforming model-free #ReinforcementLearning methods like PPO in sample efficiency. The key enabler is to combine a high-fidelity, non-differentiable simulator for forward dynamics with a simplified surrogate model for gradient backpropagation. Our framework enables learning quadruped walking in simulation in minutes without parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills on challenging terrains in minutes. This work highlights one of the first successful real-world applications of differentiable simulation for quadruped robots, offering a compelling alternative to traditional RL methods. Kudos to Yunlong Song! UZH Science University of Zurich UZH Space Hub UZH IfI European Research Council (ERC) Massachusetts Institute of Technology (MIT)MechE

Davide Scaramuzza

15,533 views • 1 year ago

Here's a tutorial in case you want to use VPN. Thanks to a friend for helping with the video :D

Here's a tutorial in case you want to use VPN. Thanks to a friend for helping with the video :D

Star Rail Universe

26,570 views • 3 months ago

Here's a tutorial in case you want to use VPN. Thanks to a friend for helping with the video :D

Here's a tutorial in case you want to use VPN. Thanks to a friend for helping with the video :D

Genshin Universe

15,981 views • 3 months ago

🔥Maximize your experience with #MistTrack ! Our comprehensive tutorial video is now available to show you how to effectively utilize the platform. Get started👉

🔥Maximize your experience with #MistTrack ! Our comprehensive tutorial video is now available to show you how to effectively utilize the platform. Get started👉

MistTrack🕵️

54,825 views • 3 years ago

🤖 AgentVerse 🪐 with a Gradio demo github: AgentVerse offers a versatile framework that streamlines the process of creating custom multi-agent environments for large language models (LLMs). Designed to facilitate swift development and customization with minimal effort, our framework empowers researchers to concentrate on their research, rather than being bogged down by implementation details.

🤖 AgentVerse 🪐 with a Gradio demo github: AgentVerse offers a versatile framework that streamlines the process of creating custom multi-agent environments for large language models (LLMs). Designed to facilitate swift development and customization with minimal effort, our framework empowers researchers to concentrate on their research, rather than being bogged down by implementation details.

AK

107,791 views • 2 years ago

🤯 Just build Fortnite in React ... With the new releases of r3/timeline and r3/viverse, building a character controller with Three.js is now easier then ever. Checkout the link below for a full tutorial on to build the demo in the video.

🤯 Just build Fortnite in React ... With the new releases of r3/timeline and r3/viverse, building a character controller with Three.js is now easier then ever. Checkout the link below for a full tutorial on to build the demo in the video.

Bela Bohlender

103,527 views • 7 months ago

As part of our goal to make MLX a great research tool, we're expanding support to new languages like Swift and C, making experimentation on Apple silicon easier for ML researchers. Video generating text with Mistral 7B and MLX Swift 👇 MLX is an array framework for machine learning research on Apple silicon. MLX is intended for research and not for production deployment of models in apps.

As part of our goal to make MLX a great research tool, we're expanding support to new languages like Swift and C, making experimentation on Apple silicon easier for ML researchers. Video generating text with Mistral 7B and MLX Swift 👇 MLX is an array framework for machine learning research on Apple silicon. MLX is intended for research and not for production deployment of models in apps.

Awni Hannun

88,745 views • 2 years ago

🎥New tutorial video is here! 🎉 👇Follow along to use our features more fluently. We've added the video to our GitBook docs (now with a Chinese version🇨🇳 too): #btc #Ordinals #BRC20 #sats

🎥New tutorial video is here! 🎉 👇Follow along to use our features more fluently. We've added the video to our GitBook docs (now with a Chinese version🇨🇳 too): #btc #Ordinals #BRC20 #sats

BSwap-Brc20 DEX

11,331 views • 2 years ago

How to rig a tail to animate waving with just one controller? With drivers: I used this expression: sin(phase * 3 - 0.5 * length) * amp more details in video #b3d #rigging #tutorial

How to rig a tail to animate waving with just one controller? With drivers: I used this expression: sin(phase * 3 - 0.5 * length) * amp more details in video #b3d #rigging #tutorial

iltaen

34,795 views • 3 months ago

Today we've launched our Outcomes Framework which shows the commitments that guide us as we seek to improve the health and wealth of the nation through research. The Framework is a statement of our intended outcomes and a tool for tracking our progress.

Today we've launched our Outcomes Framework which shows the commitments that guide us as we seek to improve the health and wealth of the nation through research. The Framework is a statement of our intended outcomes and a tool for tracking our progress.

National Institute for Health and Care Research

12,247 views • 2 years ago

Is there a crisis at the southern border? More like 10 of them. Our new video explains how a perfect storm of circumstances led to a breakdown in immigration policy.

Is there a crisis at the southern border? More like 10 of them. Our new video explains how a perfect storm of circumstances led to a breakdown in immigration policy.

Kite & Key Media

15,912,650 views • 1 year ago

Connecting Claude with web search is magic! Finally got the chance to play with MCP servers. I hooked up Claude Desktop with GitHub & Brave web search MCP servers for staying up to date with AI news. I also asked Claude to create a new GitHub repo and log the news there. This exceeded my expectations. Check the full video (at 2x) for the first demo. Let me know if you are interested in a full tutorial on how to set up the MCP servers.

Connecting Claude with web search is magic! Finally got the chance to play with MCP servers. I hooked up Claude Desktop with GitHub & Brave web search MCP servers for staying up to date with AI news. I also asked Claude to create a new GitHub repo and log the news there. This exceeded my expectations. Check the full video (at 2x) for the first demo. Let me know if you are interested in a full tutorial on how to set up the MCP servers.

elvis

41,804 views • 1 year ago

I'm excited to share our new work Diffusion as Shader (DaS), a versatile controllable video generation method for various tasks: object manipulation, camera control, mesh-to-video, and motion transfer. Project page: Github:

I'm excited to share our new work Diffusion as Shader (DaS), a versatile controllable video generation method for various tasks: object manipulation, camera control, mesh-to-video, and motion transfer. Project page: Github:

Yuan Liu

33,363 views • 1 year ago

Here’s the full video: Somalia’s former Petroleum DG, Eng. Karar Dhoomey, explains that the Türkiye deal is a framework

Here’s the full video: Somalia’s former Petroleum DG, Eng. Karar Dhoomey, explains that the Türkiye deal is a framework

HornPulse

67,651 views • 1 year ago

very quick tutorial for how to calibrate your ps5 controller sticks. circularity sweet spot is different for every controller but 6-9% seems to be consistent for most , you want the colors to be even so you get a consistent flick. for deadzone ideally is obviously 0 but due to sticks being controlled by a spring it will eventually get a slight drift. the websites used in this video are &

very quick tutorial for how to calibrate your ps5 controller sticks. circularity sweet spot is different for every controller but 6-9% seems to be consistent for most , you want the colors to be even so you get a consistent flick. for deadzone ideally is obviously 0 but due to sticks being controlled by a spring it will eventually get a slight drift. the websites used in this video are &

🎮👑

114,731 views • 5 months ago

There are three areas where every developer should focus on right now: 1. Learning how to use AI as a copilot 2. Learning how to be a copilot for AI 3. Learning how to build tooling for the future of AI Reality check: AI is here. AI isn't going anywhere. You either adapt, or somebody who does will send you to a shoelace factory for the rest of your career. I recorded a video to discuss this.

There are three areas where every developer should focus on right now: 1. Learning how to use AI as a copilot 2. Learning how to be a copilot for AI 3. Learning how to build tooling for the future of AI Reality check: AI is here. AI isn't going anywhere. You either adapt, or somebody who does will send you to a shoelace factory for the rest of your career. I recorded a video to discuss this.

Santiago

28,690 views • 1 year ago

Gemini Omni explains science with video ...... holy shit Demis Hassabis Logan Kilpatrick thanks a lot for this , now every student will get there custom video for there topic of science and math I am so happy like while typing i want to see all of your reaction to this , this is new era

Gemini Omni explains science with video ...... holy shit Demis Hassabis Logan Kilpatrick thanks a lot for this , now every student will get there custom video for there topic of science and math I am so happy like while typing i want to see all of your reaction to this , this is new era

Chetaslua

256,171 views • 1 month ago

TWIST: a real-time teleoperation system for humanoid robots to mimic whole-body motions. Reference motion data is generated by retargeting human motion-capture data to the robot. Then, the controller is trained in simulation using reinforcement learning and behavior cloning.

TWIST: a real-time teleoperation system for humanoid robots to mimic whole-body motions. Reference motion data is generated by retargeting human motion-capture data to the robot. Then, the controller is trained in simulation using reinforcement learning and behavior cloning.

The Humanoid Hub

46,715 views • 1 year ago

Hip Hop Caucus is ending 2022 with a bang—and we made a video that highlights the monumental impact we were able to have in 2022 thanks to your support! 😎🙌🏾 See ya in 2023! ✊🏾

Hip Hop Caucus is ending 2022 with a bang—and we made a video that highlights the monumental impact we were able to have in 2022 thanks to your support! 😎🙌🏾 See ya in 2023! ✊🏾

Hip Hop Caucus 🏁

141,636 views • 3 years ago