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.

DeepMind training bipedal robots to play soccer using deep reinforcement learning.

MachinePix

207,009 subscribers

16,986,259 views • 2 years ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

10 Comments

𝕏 Posting2 years ago

it'll remember this day 😂 "push me around one more time"

Bharat Vasan 🇺🇸2 years ago

if skynet takes over, videos like this will be why

MRG🥍🏀🏈🏉🤿👨‍🌾2 years ago

Sure, but when I use "deep reinforcement learning" on my kid they call Child Services.

Joe Ward ▲2 years ago

What happens when the AI realizes that finger breaking might be beneficial to producing positive outcomes in soccer play, under some conditions?

{ Luh }2 years ago

Why she keep knocking poor buddy over like that He’s trying his best!!

Paul Leavitt2 years ago

Please be nice to the baby robot, he’s gonna turn on us someday. 🥺

Tom Bielecki2 years ago

I made one of those! —he’s 2

@stevenvoiceover2 years ago

Is it wrong that I’m feeling sorry for this robot? 🙄😂

trøut2 years ago

soon

G.I Joe2 years ago

Imagine a few of these running around cleaning the house

Related Videos

Miniature humanoid robots master a simplified soccer game using Deep Reinforcement Learning, showcasing surprising agility, quick fall recovery, and strategic understanding of the game. #engineering

Miniature humanoid robots master a simplified soccer game using Deep Reinforcement Learning, showcasing surprising agility, quick fall recovery, and strategic understanding of the game. #engineering

Interesting Engineering

75,943 views • 3 years ago

Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning investigated the application of Deep Reinforcement Learning (Deep RL) for low-cost, miniature humanoid hardware in a dynamic environment, showing the method can synthesize sophisticated and safe movement skills making up complex behavioral strategies in a simplified one-versus-one (1v1) soccer game abs: project page:

Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning investigated the application of Deep Reinforcement Learning (Deep RL) for low-cost, miniature humanoid hardware in a dynamic environment, showing the method can synthesize sophisticated and safe movement skills making up complex behavioral strategies in a simplified one-versus-one (1v1) soccer game abs: project page:

AK

293,352 views • 3 years ago

Google's DeepMind trained robots how to play soccer. Here's what I found:

Google's DeepMind trained robots how to play soccer. Here's what I found:

Alex Banks

253,772 views • 3 years ago

Training AI to Play Pokemon with Reinforcement Learning by Peter Whidden github: youtube:

Training AI to Play Pokemon with Reinforcement Learning by Peter Whidden github: youtube:

AK

837,796 views • 2 years ago

NEWS: DeepMind teaches robots to play soccer. In a paper released today Google's DeepMind details how they used Deep Reinforcement Learning (Deep RL) to train low-cost, miniature humanoid robots in dynamic environments, allowing them to play a simplified one-versus-one (1v1) soccer game. The robots, equipped with 20 actuated joints, were initially trained in simulation using the MuJoCo physics engine. Through this training, they learned robust and dynamic movement skills like rapid fall recovery, walking, turning, and kicking. The robots seamlessly transitioned between these skills, even surpassing expectations, and developed a basic strategic understanding of the game. During matches, the trained robots demonstrated agile skills such as turning, kicking moving balls, and dynamic defensive blocking. They quickly combined these skills, showcasing their adaptability and outperforming scripted baselines. The robots walked faster, got up quicker, and kicked faster than their counterparts. Individual skills were first trained in isolation within the simulation environment, and then composed in a self-play setting. The robots successfully transferred these skills to real-world scenarios. Full paper can be found here:

NEWS: DeepMind teaches robots to play soccer. In a paper released today Google's DeepMind details how they used Deep Reinforcement Learning (Deep RL) to train low-cost, miniature humanoid robots in dynamic environments, allowing them to play a simplified one-versus-one (1v1) soccer game. The robots, equipped with 20 actuated joints, were initially trained in simulation using the MuJoCo physics engine. Through this training, they learned robust and dynamic movement skills like rapid fall recovery, walking, turning, and kicking. The robots seamlessly transitioned between these skills, even surpassing expectations, and developed a basic strategic understanding of the game. During matches, the trained robots demonstrated agile skills such as turning, kicking moving balls, and dynamic defensive blocking. They quickly combined these skills, showcasing their adaptability and outperforming scripted baselines. The robots walked faster, got up quicker, and kicked faster than their counterparts. Individual skills were first trained in isolation within the simulation environment, and then composed in a self-play setting. The robots successfully transferred these skills to real-world scenarios. Full paper can be found here:

Dave Lee

372,039 views • 3 years ago

Watch Gemini 2.5 Pro implement a landmark Google DeepMind research paper. 🕹️ It codes the reinforcement learning algorithm, visualizes the training live and even debugs errors. ↓

Watch Gemini 2.5 Pro implement a landmark Google DeepMind research paper. 🕹️ It codes the reinforcement learning algorithm, visualizes the training live and even debugs errors. ↓

Google DeepMind

369,902 views • 1 year ago

Introducing P1: the cutest and toughest bipedal robot mastering AI reinforcement learning to conquer wild terrains while shrugging off kicks and sticks! 🤖

Introducing P1: the cutest and toughest bipedal robot mastering AI reinforcement learning to conquer wild terrains while shrugging off kicks and sticks! 🤖

Interesting Engineering

22,055 views • 1 year ago

Trying to teach my new model to run using reinforcement learning (Hyfydy + depRL)

Trying to teach my new model to run using reinforcement learning (Hyfydy + depRL)

Thomas Geijtenbeek

975,809 views • 1 year ago

Figure 02 robots are now walking with a smoother gait and increased speed, thanks to sim-to-real reinforcement learning.

Figure 02 robots are now walking with a smoother gait and increased speed, thanks to sim-to-real reinforcement learning.

The Humanoid Hub

146,868 views • 1 year ago

Our CoRL 2024 paper shows Reinforcement Learning can allow robots to learn skills via real-world practice, without any demonstrations or simulation engineering. Rewards are provided using language/vision models, and mobility of robots enables autonomous exploration. 1/N

Our CoRL 2024 paper shows Reinforcement Learning can allow robots to learn skills via real-world practice, without any demonstrations or simulation engineering. Rewards are provided using language/vision models, and mobility of robots enables autonomous exploration. 1/N

Russell Mendonca

38,454 views • 1 year ago

Its nice to see the recent excitement around Q-learning! Here's a clip from a video all about Deep Reinforcement Learning, including Deep Q Learning #machinelearning #ai

Its nice to see the recent excitement around Q-learning! Here's a clip from a video all about Deep Reinforcement Learning, including Deep Q Learning #machinelearning #ai

Steven Brunton

164,997 views • 2 years ago

LimX Dynamics P1 is a versatile bipedal robot equipped with reinforcement learning capabilities, enabling it to master challenging terrains such as forests and ditches while maintaining its balance.

LimX Dynamics P1 is a versatile bipedal robot equipped with reinforcement learning capabilities, enabling it to master challenging terrains such as forests and ditches while maintaining its balance.

Science girl

3,295,229 views • 2 years ago

When will a team of robots be able to beat a professional soccer team in the game of soccer? 🤔 Many in robotics have thought that it would take at least 15-20 years for robots to reach that point. But today 60 Minutes showed some of Google Deep Mind's work on robots playing soccer. Robots are using neural nets to learn how to walk, kick and play soccer, and they're improving very rapidly. It might be a lot sooner than most people think before we have robots able to beat professional athletes at their own sports.

When will a team of robots be able to beat a professional soccer team in the game of soccer? 🤔 Many in robotics have thought that it would take at least 15-20 years for robots to reach that point. But today 60 Minutes showed some of Google Deep Mind's work on robots playing soccer. Robots are using neural nets to learn how to walk, kick and play soccer, and they're improving very rapidly. It might be a lot sooner than most people think before we have robots able to beat professional athletes at their own sports.

Dave Lee

177,070 views • 3 years ago

i trained a reinforcement learning agent to play diablo 1 in a few hours with pufferlib

i trained a reinforcement learning agent to play diablo 1 in a few hours with pufferlib

Dan Advantage

41,954 views • 7 months ago

Reinforcement learning experiments continues. Still a bit janky, but it is clearly learning something. This is about an hour into training.

Reinforcement learning experiments continues. Still a bit janky, but it is clearly learning something. This is about an hour into training.

Dennis Gustafsson

22,362 views • 10 months 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

Frontier models that use reasoning to "think" during inference are generating 5X more AI tokens per year. ✨ "Inference is now a thinking process. And in order to teach AI how to think, reinforcement learning and very significant computation was introduced into post-training," said NVIDIA CEO Jensen Huang in a recent keynote at #CES2026. Reinforcement learning is increasing computation demands across all AI scaling laws: pre-training, post-training, and test-time scaling. Learn more about reinforcement learning and AI scaling ➡️

Frontier models that use reasoning to "think" during inference are generating 5X more AI tokens per year. ✨ "Inference is now a thinking process. And in order to teach AI how to think, reinforcement learning and very significant computation was introduced into post-training," said NVIDIA CEO Jensen Huang in a recent keynote at #CES2026. Reinforcement learning is increasing computation demands across all AI scaling laws: pre-training, post-training, and test-time scaling. Learn more about reinforcement learning and AI scaling ➡️

NVIDIA AI Infrastructure

13,329 views • 6 months ago

🔥Exciting news 🤖 Our latest research by David Hoeller, Nikita Rudin, Eris in Science Robotics unlocks new achievements: Unprecedented agility in quadrupedal robots, mastering locomotion, navigation, and perception through deep reinforcement learning! NVIDIA Robotics

🔥Exciting news 🤖 Our latest research by David Hoeller, Nikita Rudin, Eris in Science Robotics unlocks new achievements: Unprecedented agility in quadrupedal robots, mastering locomotion, navigation, and perception through deep reinforcement learning! NVIDIA Robotics

Robotic Systems Lab

171,798 views • 2 years ago