AlphaStar - Grandmaster Level in StarCraft II Using Multi-Agent Reinforcement Learning
Offered By: Yannic Kilcher via YouTube
Course Description
Overview
Explore DeepMind's groundbreaking AI agent AlphaStar, which achieved Grandmaster-level skill in StarCraft II using multi-agent reinforcement learning. Dive into the complex world of competitive esports AI as this 37-minute video breaks down the innovative techniques behind AlphaStar's success. Learn about the League Training method that propelled the agent to top-tier performance, surpassing 99.8% of officially ranked human players. Examine how AlphaStar tackles the raw complexity and multi-agent challenges of StarCraft II, setting a new benchmark in artificial intelligence for real-time strategy games. Gain insights into the general-purpose learning methods employed, including deep neural networks and a diverse league of continually adapting strategies. Understand the significance of this achievement in the context of AI research and its potential applications to other complex domains beyond gaming.
Syllabus
AlphaStar: Grandmaster level in StarCraft II using multi-agent reinforcement learning
Taught by
Yannic Kilcher
Related Courses
Reinforcement Learning FoundationsLinkedIn Learning Toward Generalizable Embodied AI for Machine Autonomy
Bolei Zhou via YouTube What Does Machine Learning Offer Game Theory - And Vice Versa
Simons Institute via YouTube What Is the Statistical Complexity of Reinforcement Learning?
Simons Institute via YouTube Multi-Agent Reinforcement Learning - Part II
Simons Institute via YouTube