Reinforcement Learning: AlphaGo

Graphics in 5 Minutes · Advanced ·📰 AI News & Updates ·2y ago
How AlphaGo works, based on Reinforcement Learning. Part 2 of RL from scratch series. https://youtu.be/vXtfdGphr3c 0:00 - intro 0:06 - how to play Go 0:21 - introducing alphaGo 0:46 - analyzing expert games 2:17 - training an expert policy 2:47 - value functions 4:05 - search trees 5:42 - reinforcement learning 6:17 - alphaGo's value function 7:47 - alphaZero
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related AI Lessons

Meta Is in Crisis, Google Search’s Makeover, and AI Gets Booed by Graduates
Learn about Meta's crisis, Google Search's makeover, and AI's backlash from graduates, and how these events impact the tech industry
Wired AI
What I Do Between Biotech Jobs, Part 1: The 20-Line Script That Outsmarted an AI
Learn how a 20-line script outsmarted an AI in biotech, and discover the potential of creative problem-solving in the industry
Medium · AI
Spotify and Universal Music strike deal allowing fan-made AI covers and remixes
Spotify and Universal Music partner to allow fan-made AI covers and remixes, with revenue sharing for artists
TechCrunch AI
Google Just Turned Search Into Something It Has Never Been Before
Google Search has undergone a significant transformation after 25 years, leveraging AI to change its core functionality
Medium · AI

Chapters (10)

intro
0:06 how to play Go
0:21 introducing alphaGo
0:46 analyzing expert games
2:17 training an expert policy
2:47 value functions
4:05 search trees
5:42 reinforcement learning
6:17 alphaGo's value function
7:47 alphaZero
Up next
OpenAI: $2M in tokens to every YC company in the spring and summer batches.
Y Combinator
Watch →