How to Train a Robot Arm - A New Method
In this video, I discuss the paper "Asymmetric self-play for automatic goal discovery in robotic manipulation," which describes an effective reinforcement learning method for training robots in simulation by using self-play (robot vs robot). One robot's goal is to move the objects around in a way that it thinks the other robot won't be able to copy, and then the other robot receives this end state as its goal and it tries to move the objects around to match it.
My 3 favorite things about this method are:
1. You get an auto-generated curriculum for the robot to learn from.
2. You know that each task is solvable since a robot created it.
3. You automatically get a demonstration of each task that can be learned from.
📄 The Paper:
https://arxiv.org/abs/2101.04882
Authors (OpenAI):
Matthias Plappert, Raul Sampedro, Tao Xu, Ilge Akkaya, Vineet Kosaraju, Peter Welinder, Ruben D'Sa, Arthur Petron, Henrique Ponde de Oliveira Pinto, Alex Paino, Hyeonwoo Noh, Lilian Weng, Qiming Yuan, Casey Chu, Wojciech Zaremba
Connect with me:
🐦 Twitter - https://twitter.com/elliotwaite
💬 Discord - https://discord.gg/cdQhRgw
📷 Instagram - https://www.instagram.com/elliotwaite
💼 LinkedIn - https://www.linkedin.com/in/elliotwaite
Видео How to Train a Robot Arm - A New Method канала Elliot Waite
My 3 favorite things about this method are:
1. You get an auto-generated curriculum for the robot to learn from.
2. You know that each task is solvable since a robot created it.
3. You automatically get a demonstration of each task that can be learned from.
📄 The Paper:
https://arxiv.org/abs/2101.04882
Authors (OpenAI):
Matthias Plappert, Raul Sampedro, Tao Xu, Ilge Akkaya, Vineet Kosaraju, Peter Welinder, Ruben D'Sa, Arthur Petron, Henrique Ponde de Oliveira Pinto, Alex Paino, Hyeonwoo Noh, Lilian Weng, Qiming Yuan, Casey Chu, Wojciech Zaremba
Connect with me:
🐦 Twitter - https://twitter.com/elliotwaite
💬 Discord - https://discord.gg/cdQhRgw
📷 Instagram - https://www.instagram.com/elliotwaite
💼 LinkedIn - https://www.linkedin.com/in/elliotwaite
Видео How to Train a Robot Arm - A New Method канала Elliot Waite
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![Policy Gradient Theorem Explained - Reinforcement Learning](https://i.ytimg.com/vi/cQfOQcpYRzE/default.jpg)
![Cellular Automata: Rule 110 + Conway’s Game of Life](https://i.ytimg.com/vi/P2uhhAXd7PI/default.jpg)
![PyTorch Hooks Explained - In-depth Tutorial](https://i.ytimg.com/vi/syLFCVYua6Q/default.jpg)
![Sample efficient reinforcement learning with SAC on Kuka robot](https://i.ytimg.com/vi/FuLbkhhJOD4/default.jpg)
![Precise Robot Manipulation with Never-Before-Seen Objects](https://i.ytimg.com/vi/l9U8X6I1vow/default.jpg)
![Why Does This Pattern Emerge?](https://i.ytimg.com/vi/sgXjzlC1d_4/default.jpg)
![Best PhD Apps | 10+ essential apps for every PhD student](https://i.ytimg.com/vi/1A9GJjmDyKI/default.jpg)
![How Do Robot Arms Work?](https://i.ytimg.com/vi/QXkBMs30JAs/default.jpg)
![Neural Network Learns to Play Snake](https://i.ytimg.com/vi/zIkBYwdkuTk/default.jpg)
![Clock Assembly Automation with Robot Arms](https://i.ytimg.com/vi/zHozy5axc1c/default.jpg)
![Solidworks - Robotic Arm 💥](https://i.ytimg.com/vi/9p3E0MtCDho/default.jpg)
![This Robot Drives on Wobbly Wheels (but HOW?)](https://i.ytimg.com/vi/r9ZTWQtzk2Y/default.jpg)
![How computers learn to recognize objects instantly | Joseph Redmon](https://i.ytimg.com/vi/Cgxsv1riJhI/default.jpg)
![Robot Exhibition. Highlights from CES 2020. The Coolest Robots and Incredible Gadgets!](https://i.ytimg.com/vi/QfhQkLyJnsE/default.jpg)
![How to Run PyTorch Models in the Browser With ONNX.js](https://i.ytimg.com/vi/Vs730jsRgO8/default.jpg)
![Universal Robots - Easy Pallet Routine](https://i.ytimg.com/vi/jkAY1VNi_L0/default.jpg)
![TACTILE TELEROBOT - WIRED DEMO](https://i.ytimg.com/vi/XdERNxRqQ4Q/default.jpg)
![SproutLife - Simulating the evolution of complex life](https://i.ytimg.com/vi/H2tXsnzVTaw/default.jpg)
![A Robot Arm That's Easy to Program](https://i.ytimg.com/vi/vAiuwpHPeqk/default.jpg)
![Deep Reinforcement Learning for Robotic Arm Manipulation](https://i.ytimg.com/vi/ub4ZyegbTSw/default.jpg)