From AlphaGo Onward: What Have We Actually Seen

From AlphaGo Onward: What Have We Actually Seen — a retrospective on AlphaGo's Go career, from the 5–0 win over Fan Hui in October 2015 through Lee Sedol, Master, Ke Jie, and AlphaGo Zero in October 2017. The article separates three ideas — the verifier, search, and learning — and draws six lessons from the process alone: evaluation functions can be learned; the target must be set right first; search is the teacher and the network the student; human joseki is a feasible solution, not the optimal one; superhuman holds only within a domain; and the human job has changed from playing Go to reading Go. Both the English and Chinese versions are included in this record. Research and drafting were assisted by AI. All analysis, judgments, and the final text are the author's own.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-12
DOI
https://doi.org/10.5281/zenodo.24177003
Primary Topic
Artificial Intelligence in Games
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

From AlphaGo Onward: What Have We Actually Seen

Xuefeng Yang
Zenodo (CERN European Organization for Nuclear Research)
Artificial Intelligence in Games
preprint

From AlphaGo Onward: What Have We Actually Seen

Xuefeng Yang
preprint en

Abstract

From AlphaGo Onward: What Have We Actually Seen — a retrospective on AlphaGo's Go career, from the 5–0 win over Fan Hui in October 2015 through Lee Sedol, Master, Ke Jie, and AlphaGo Zero in October 2017. The article separates three ideas — the verifier, search, and learning — and draws six lessons from the process alone: evaluation functions can be learned; the target must be set right first; search is the teacher and the network the student; human joseki is a feasible solution, not the optimal one; superhuman holds only within a domain; and the human job has changed from playing Go to reading Go. Both the English and Chinese versions are included in this record. Research and drafting were assisted by AI. All analysis, judgments, and the final text are the author's own.

Zenodo (CERN European Organization for Nuclear Research)
Artificial Intelligence in Games
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon