YoVDO

AMP- Adversarial Motion Priors for Stylized Physics-Based Character Control

Offered By: Yannic Kilcher via YouTube

Tags

Reinforcement Learning Courses Imitation Learning Courses

Course Description

Overview

Explore a comprehensive video explanation of the paper "AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control". Delve into the innovative approach of combining goal-achieving reinforcement learning with imitation learning using GANs to create realistic character movements. Learn about the problem statement, reward signals, motion priors from GANs, algorithm overview, reward engineering, and experimental results. Gain insights into how this method produces high-quality motions comparable to state-of-the-art tracking-based techniques while accommodating large datasets of unstructured motion clips. Discover how the system automatically composes disparate skills without requiring a high-level motion planner or task-specific annotations.

Syllabus

- Intro & Overview
- Problem Statement
- Reward Signals
- Motion Prior from GAN
- Algorithm Overview
- Reward Engineering & Experimental Results
- Conclusion & Comments


Taught by

Yannic Kilcher

Related Courses

Computational Neuroscience
University of Washington via Coursera
Reinforcement Learning
Brown University via Udacity
Reinforcement Learning
Indian Institute of Technology Madras via Swayam
FA17: Machine Learning
Georgia Institute of Technology via edX
Introduction to Reinforcement Learning
Higher School of Economics via Coursera