正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Reinforcement Learning for Robotics Part 2: Train a Balance Bot with PPO with Shawn Hymel Full video:

maker.io

57,081 subscribers

14,017 次观看 • 3 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

1 条评论

Mandar Wagh 的头像

Mandar Wagh2 天前

@ShawnHymel the number that decides whether it transfers: a 30 cm inverted pendulum has a time constant of sqrt(L/g), about 175 ms, and the error grows by a factor of e over that. sim has zero latency, so every millisecond of real sensor and actuator delay comes out of that budget.

相关视频

Years of reinforcement learning in one clip: Atlas pulls off a cartwheel-to-backflip combo with perfect balance.

Years of reinforcement learning in one clip: Atlas pulls off a cartwheel-to-backflip combo with perfect balance.

Big Brain AI

91,653 次观看 • 4 个月前

notes on my reinforcement learning for rocket league side quest so far: i’m in awe that — we live in a time where i can train a reinforcement learning model that is learning to play rocket league and at the same time, test community made rocket league models. where, both processes can run in parallel & locally on my hardware. in the game window — i’m playing against a reinforcement learning ppo model that’s running locally. and in the vscode window — i’m training my own reinforcement learning model (just a dummy example script for now). also, the rocket league botting ecosystem has incredible tooling where you can test out different reinforcement learning ideas against other algorithms + it has great documentation for linux and windows. it’s honestly a beautiful hidden gem of a rabbit hole :)

notes on my reinforcement learning for rocket league side quest so far: i’m in awe that — we live in a time where i can train a reinforcement learning model that is learning to play rocket league and at the same time, test community made rocket league models. where, both processes can run in parallel & locally on my hardware. in the game window — i’m playing against a reinforcement learning ppo model that’s running locally. and in the vscode window — i’m training my own reinforcement learning model (just a dummy example script for now). also, the rocket league botting ecosystem has incredible tooling where you can test out different reinforcement learning ideas against other algorithms + it has great documentation for linux and windows. it’s honestly a beautiful hidden gem of a rabbit hole :)

naklecha

46,795 次观看 • 1 年前

Train your avatars to interact with 3D scenes. We use adversarial imitation learning and reinforcement learning to train physically-simulated characters that perform scene interaction tasks in a natural and life-like manner. Today at #SIGGRAPH2023.

Train your avatars to interact with 3D scenes. We use adversarial imitation learning and reinforcement learning to train physically-simulated characters that perform scene interaction tasks in a natural and life-like manner. Today at #SIGGRAPH2023.

Michael Black

46,237 次观看 • 3 年前

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

3,295,511 次观看 • 2 年前

how to learn something from the first principles (with a live demo on reinforcement learning)

how to learn something from the first principles (with a live demo on reinforcement learning)

Ajeet ( opensox.ai )

234,216 次观看 • 4 个月前

Get certified with our new course, "Introduction to Edge AI." This free 10-part series will help you become familiar with edge computing and #MachineLearning, taught by Shawn Hymel. We’re excited to see what you will build! Start here 👉

Get certified with our new course, "Introduction to Edge AI." This free 10-part series will help you become familiar with edge computing and #MachineLearning, taught by Shawn Hymel. We’re excited to see what you will build! Start here 👉

Edge Impulse

14,534 次观看 • 2 年前

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 次观看 • 9 个月前

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

293,474 次观看 • 3 年前

In order for robots to be deployed in the real world, performing tasks of real value, they must be reliable. Unfortunately, even more, most robotic demos work maybe 70-80% of the time at best. The way to get better reliability is to do real-world reinforcement learning: having the robot teach itself how to perform the task up to a high level of success. The key to doing this is to start with a core of expert human data, use that to train a policy then iteratively improve it, until finally finishing with on-policy reinforcement learning. Kun Lei talks through a unified framework for imitation and reinforcement learning based on PPO, which enables this improvement process. In this episode, Kun Lei explains the theory behind his reinforcement learning method and how it allowed his robot to run in a shopping mall juicing oranges for seven hours at a time, among experiments on a wide variety of tasks and embodiments. Watch episode 58 of RoboPapers now, hosted by Michael Cho - Rbt/Acc and Chris Paxton!

In order for robots to be deployed in the real world, performing tasks of real value, they must be reliable. Unfortunately, even more, most robotic demos work maybe 70-80% of the time at best. The way to get better reliability is to do real-world reinforcement learning: having the robot teach itself how to perform the task up to a high level of success. The key to doing this is to start with a core of expert human data, use that to train a policy then iteratively improve it, until finally finishing with on-policy reinforcement learning. Kun Lei talks through a unified framework for imitation and reinforcement learning based on PPO, which enables this improvement process. In this episode, Kun Lei explains the theory behind his reinforcement learning method and how it allowed his robot to run in a shopping mall juicing oranges for seven hours at a time, among experiments on a wide variety of tasks and embodiments. Watch episode 58 of RoboPapers now, hosted by Michael Cho - Rbt/Acc and Chris Paxton!

RoboPapers

18,813 次观看 • 8 个月前

🚀 🔥 AgiBot deploys Real-World Reinforcement Learning (RW-RL) in industrial robotics with Longcheer Technology. Robots now learn new skills in tens of MINUTES (not weeks), adapt to variations autonomously, and reconfigure flexibly, solving rigid automation pain points in precision manufacturing. A giant leap of intelligent automation for precision manufacturing! #AgiBot #ReinforcementLearning #RealWorldRL #Robotics #AI #IndustrialRobotics

🚀 🔥 AgiBot deploys Real-World Reinforcement Learning (RW-RL) in industrial robotics with Longcheer Technology. Robots now learn new skills in tens of MINUTES (not weeks), adapt to variations autonomously, and reconfigure flexibly, solving rigid automation pain points in precision manufacturing. A giant leap of intelligent automation for precision manufacturing! #AgiBot #ReinforcementLearning #RealWorldRL #Robotics #AI #IndustrialRobotics

AGIBOT

131,860 次观看 • 10 个月前

In 2011, Dr. Guero designed the worlds first custom-made bicycling robotics model with special joints and a gyroscope for balance.

In 2011, Dr. Guero designed the worlds first custom-made bicycling robotics model with special joints and a gyroscope for balance.

HOW THINGS WORK

118,278 次观看 • 1 年前

We're excited to share an early video of Apollo integrating with NVIDIA Robotics' new general-purpose foundation model for robot learning, Project GR00T! #NVIDIARobotics #GTC24

We're excited to share an early video of Apollo integrating with NVIDIA Robotics' new general-purpose foundation model for robot learning, Project GR00T! #NVIDIARobotics #GTC24

Apptronik

52,859 次观看 • 2 年前

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 次观看 • 1 年前

Robot policies must be both reliable and highly capable to be useful; the best way to achieve this level of performance is with reinforcement learning. However, for reinforcement learning you are usually stuck between two difficult options: reinforcement in the real world is often risky and expensive, while reinforcement learning in a traditional simulator takes a lot of engineering work and has a persistent sim-to-real gap. What if instead you could train your robot purely in a world model? RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning. Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

Robot policies must be both reliable and highly capable to be useful; the best way to achieve this level of performance is with reinforcement learning. However, for reinforcement learning you are usually stuck between two difficult options: reinforcement in the real world is often risky and expensive, while reinforcement learning in a traditional simulator takes a lot of engineering work and has a persistent sim-to-real gap. What if instead you could train your robot purely in a world model? RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning. Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

RoboPapers

38,772 次观看 • 3 个月前

The teaser video for our latest Science Robotics paper. Full video will be out soon. Paper: Attention-based map encoding for learning generalized legged locomotion Link:

The teaser video for our latest Science Robotics paper. Full video will be out soon. Paper: Attention-based map encoding for learning generalized legged locomotion Link:

Junzhe (JJ) He

13,162 次观看 • 1 年前

Learning robust, general-purpose reward functions for robotics unlocks many potential applications, like on-robot reinforcement learning or dataset validation. However, there’s a question of how to actually train such reward functions. Training success/failure prediction leads to ambiguous signals partway through a demonstration — it’s hard to measure progress — making the method unsuitable for reinforcement learning, among other things. Predicting progress, on the other hand, does not give a good way of using failure data. So why not do both? Robometer combines both progress and preference supervision, resulting in a stable, scalable, and highly general reward learning approach. Anthony Liang Yigit Korkmaz and Jesse Zhang join us to tell us more. Watch Episode #84 of RoboPapers, with Chris Paxton and Jiafei Duan today!

Learning robust, general-purpose reward functions for robotics unlocks many potential applications, like on-robot reinforcement learning or dataset validation. However, there’s a question of how to actually train such reward functions. Training success/failure prediction leads to ambiguous signals partway through a demonstration — it’s hard to measure progress — making the method unsuitable for reinforcement learning, among other things. Predicting progress, on the other hand, does not give a good way of using failure data. So why not do both? Robometer combines both progress and preference supervision, resulting in a stable, scalable, and highly general reward learning approach. Anthony Liang Yigit Korkmaz and Jesse Zhang join us to tell us more. Watch Episode #84 of RoboPapers, with Chris Paxton and Jiafei Duan today!

RoboPapers

14,751 次观看 • 3 个月前