ALPHA Timepoint is in alpha Talk to Us
Sign In Get Started
DeepMind publishes AlphaZero paper

DeepMind publishes AlphaZero paper

DeepMind researchers are presenting their breakthrough findings on AlphaZero, an AI system that mastered chess, shogi, and Go through self-play, in a modern conference room at their London offices.

Setting

A modern conference room in the DeepMind offices, London, with floor-to-ceiling windows overlooking the city. The room is equipped with a large projector screen, a sleek wooden conference table, and ergonomic chairs. Whiteboards filled with equations and diagrams line the walls.

Characters

The figures in this scene as an entity network — co-presence links everyone in the moment; speakers who trade lines are bound tighter. Turn the resolution dial to reveal depth the engine actually computed.

TNGF
SELECTED
Lead Researcher
primary
A middle-aged man with short, neatly trimmed dark hair and a clean-shaven face. He wears rectangular glasses that slightly magnify his keen, analytical eyes. His build is lean, suggesting long hours spent at a desk rather than in physical labor. His posture is upright, exuding confidence and authority.
Junior Researcher
secondary
A young man in his mid-20s with a lean build, short dark hair, and a clean-shaven face. His bright eyes dart between his notes and the presentation screen with keen interest.
Senior Scientist
secondary
A middle-aged man with short, graying hair and a neatly trimmed beard. His sharp eyes behind rectangular glasses reflect years of focused study. He has a lean build, typical of someone who spends long hours at a desk but makes time for regular exercise.
Technical Assistant
background
A young adult in their late 20s, of average height with a lean build. They have short, neatly trimmed dark brown hair and wear rectangular-framed glasses. Their movements are efficient and precise, reflecting their technical role.

Dialog

Lead Researcher So what we're seeing here is AlphaZero mastering chess, shogi, and Go purely through self-play—no human data, no pre-programmed strategies. It's like watching a child learn the rules of a game by playing against itself, but at an unprecedented scale.
Junior Researcher Wait, does that mean it... it essentially reinvented centuries of chess theory on its own?
Senior Scientist Precisely. And in less than 24 hours of training. To clarify—it didn't just replicate human knowledge. It discovered novel strategies we'd never considered.
Lead Researcher Exactly. And the implications extend far beyond games. This is a paradigm shift in how we approach machine learning—pure self-improvement without domain-specific biases.
Junior Researcher But the compute resources required... surely that limits practical applications?
Senior Scientist Building on that point—the efficiency gains are remarkable. Remember, this system achieved superhuman performance with less hardware than traditional approaches. The scaling properties are... promising.
Lead Researcher Which brings us to the real question—what happens when we apply this to problems where we don't even know the rules yet?

Chat with Characters

You've used your 3 free turns

Sign in to keep chatting with characters from this moment — unlimited turns.

Sign in to Continue
— Sign in for unlimited

Causal neighbors · 133 linked moments

E
Establishment of the MIT Artificial Intelligence Laboratory
1959 · same figure
M
Moore's Law paper published
1965 · same figure
G
GPT-3 Paper arXiv Submission
2020 · same figure
F
First point-contact transistor demonstrated
1947 · same figure
R
Release of GPT-1 paper "Improving Language Understanding by Generative Pre-Training"
2018 · same figure
R
Release of BERT (Bidirectional Encoder Representations from Transformers) Paper
2018 · same figure
R
Release of GPT-1 (Generative Pre-trained Transformer) Paper
2018 · same figure
P
Publication of BERT
2018 · same figure
G
GPT-2 Paper arXiv Submission
2019 · same figure
A
AlphaGo Zero introduced
2017 · same figure
U
UK Lighthill Report criticizes AI research
1973 · same figure
A
Association for the Advancement of Artificial Intelligence (AAAI) founded
1979 · same figure
R
Release of the Transformer paper "Attention is All You Need"
2017 · same figure
P
Publication of the AlphaGo Nature paper
2016 · same figure
F
First Silicon Transistor Developed
1954 · same figure
F
First Laser Demonstration
1960 · same figure
A
AlphaFold 2 paper published
2021 · same figure
A
AlphaGo defeats Lee Sedol
2016 · thematic
2012 London Paralympic Opening Ceremony
2012 London Paralympic Opening Ceremony
2012 · same era
2012 London Paralympic Opening Ceremony
2012 London Paralympic Opening Ceremony
2012 · precedes
W
Wedding of Prince William and Catherine Middleton
2011 · same era
D
DeepMind announces AlphaGo Zero
2017 · same era
U
United Kingdom European Union Membership Referendum (Brexit)
2016 · same era
O
Opening Ceremony of the 2012 London Olympics
2012 · same era
A
AlphaGo defeats Fan Hui
2015 · same era
Q
Queen Elizabeth Prize for Engineering awarded to Tim Berners-Lee
2013 · same era
A
AlphaGo Zero introduced
2017 · same era
D
Death of Queen Elizabeth II
2022 · same era
P
Publication of the AlphaGo Nature paper
2016 · same era
R
Release of AlphaGo Zero
2017 · same era
F
Founding of DeepMind Technologies
2010 · same era
O
Opening Ceremony of the 2012 London Olympics
2012 · same era
W
Wedding of Prince William and Catherine Middleton
2011 · same era
A
AlphaFold 2 paper published
2021 · same era
A
AlphaGo defeats Fan Hui
2015 · same era
A
AlphaGo Zero vs AlphaGo match
2017 · same era