AlphaGo - Wikipedia
webCredibility Rating
3/5
Good(3)Good quality. Reputable source with community review or editorial standards, but less rigorous than peer-reviewed venues.
Rating inherited from publication venue: Wikipedia
AlphaGo represents a landmark AI capabilities milestone, demonstrating that deep reinforcement learning can surpass human expert performance in complex domains, which is directly relevant to AI safety discussions about capability jumps, superhuman AI, and the trajectory of AI development.
Metadata
Importance: 62/100wiki pagereference
Summary
AlphaGo is DeepMind's AI program that uses Monte Carlo tree search combined with deep neural networks to play the board game Go. It became the first program to defeat professional human Go players without handicap, culminating in a 4-1 victory over world champion Lee Sedol in 2016. Its successors (AlphaGo Zero, AlphaZero, MuZero) demonstrated increasingly general and self-taught capabilities.
Key Points
- •AlphaGo was the first computer program to beat a professional Go player without handicap, a milestone considered decades away by many experts.
- •AlphaGo Zero learned entirely through self-play without human game data, demonstrating that superhuman performance can emerge without human knowledge.
- •AlphaZero generalized the approach to chess and shogi, showing the technique's broad applicability across complex strategic domains.
- •The system uses deep reinforcement learning and Monte Carlo tree search, combining neural network evaluation with tree-based planning.
- •AlphaGo's rapid capability gains illustrate how AI progress can be discontinuous and faster than anticipated, relevant to AI safety forecasting.
Cached Content Preview
HTTP 200Fetched May 11, 202666 KB
From Wikipedia, the free encyclopedia
Artificial intelligence that plays Go
This article is about a computer program. For the film, see AlphaGo (film) .
AlphaGo Developer Google DeepMind Type Computer Go software Website deepmind.com/research/highlighted-research/alphago
Part of a series on Artificial intelligence (AI)
Major goals
Artificial general intelligence
Intelligent agent
Recursive self-improvement
Planning
Computer vision
General game playing
Knowledge representation
Natural language processing
Robotics
AI safety
Approaches
Machine learning
Symbolic
Deep learning
Bayesian networks
Evolutionary algorithms
Hybrid intelligent systems
Systems integration
Open-source
AI data centers
Applications
Bioinformatics
Deepfake
Earth sciences
Finance
Generative AI
Art
Audio
Music
Government
Healthcare
Mental health
Industry
Software development
Translation
Military
Physics
Projects
Philosophy
AI alignment
Artificial consciousness
The bitter lesson
Chinese room
Friendly AI
Ethics
Existential risk
Turing test
Uncanny valley
Human–AI interaction
History
Timeline
Progress
AI winter
AI boom
AI bubble
Controversies
Deepfake pornography
Taylor Swift deepfake pornography controversy
Grok sexual deepfake scandal
Google Gemini image generation controversy
It's the Most Terrible Time of the Year
Pause Giant AI Experiments
Removal of Sam Altman from OpenAI
Statement on AI Risk
Tay (chatbot)
Théâtre D'opéra Spatial
Voiceverse NFT plagiarism scandal
Glossary
Glossary
v
t
e
AlphaGo is a computer program that plays the board game Go . [ 1 ] It was developed by the London-based DeepMind Technologies, [ 2 ] an acquired subsidiary of Google . Subsequent versions of AlphaGo became increasingly powerful, including a version that competed under the name Master . [ 3 ] After retiring from competitive play, AlphaGo Master was succeeded by an even more powerful version known as AlphaGo Zero , which was completely self-taught without learning from human games. AlphaGo Zero was then generalized into a program known as AlphaZero , which played additional games, including chess and shogi . AlphaZero has in turn been succeeded by a program known as MuZero which learns without being taught the rules.
AlphaGo and its successors use a Monte Carlo tree search algorithm to find its moves based on knowledge previously acquired by machine learning , specifically by an artificial neural network (a deep learning method) by extensive training, both from human and computer play. [ 4 ] A neural network is trained to identify the best moves and the winning percentages of these moves. This neural network improves the strength of the tree search, resulting in stronger move selection in the next iteration.
In
... (truncated, 66 KB total)Resource ID:
4c37f46e62982808 | Stable ID: sid_vWusImqhmA