LECTURE 01
FOUNDATION
INTERACTION
Common Knowledge
MODEL
M-E-P
agents + rules + payoffs
GAME THEORY
01
Klein Blue Course Interface
博弈论学习仓库
Strategic Interaction
从策略表示到机制设计，一条可执行的学习路径。
22 讲内容被整理为课程仪表盘。每一讲都能直接打开讲义、练习和配套图示，学习进度会保存在本机浏览器中。
LECTURE 05
STATIC
NASH
Best Response
FIXPOINT
BR1 x BR2
mutual best response
GAME THEORY
05
0
已完成
22
核心讲义
6
课程阶段
22
当前结果
Lecture Deck
课程讲义
LECTURE 02
FOUNDATION
PLAYERS
Strategy Space
OBJECTS
S -> u
players choose actions
GAME THEORY
02
LECTURE 03
FOUNDATION
FORMS
Tree vs Matrix
REPRESENT
G = (N,S,u)
matrix or tree view
GAME THEORY
03
LECTURE 04
STATIC
DOMINANCE
Elimination
FILTER
D -> R
delete dominated paths
GAME THEORY
04
LECTURE 05
STATIC
NASH
Best Response
FIXPOINT
BR1 x BR2
mutual best response
GAME THEORY
05
LECTURE 06
STATIC
MIXED
Probability Simplex
RANDOMIZE
p in Delta
support and indifference
GAME THEORY
06
LECTURE 07
DYNAMIC
DYNAMIC
Backward Induction
TIMELINE
V(h)
solve from the end
GAME THEORY
07
LECTURE 08
DYNAMIC
SPNE
Credible Paths
CREDIBLE
SPNE
every subgame tested
GAME THEORY
08
LECTURE 09
DYNAMIC
REPEATED
Future Shadow
REPUTATION
delta
future changes today
GAME THEORY
09
LECTURE 10
INFO
BAYES
Belief Updating
BELIEF
P(t|s)
prior becomes posterior
GAME THEORY
10
LECTURE 11
INFO
SIGNALS
Costly Messages
MESSAGE
c(e,t)
cost separates types
GAME THEORY
11
LECTURE 12
INFO
MECHANISM
IC / IR Menu
DESIGN
IC / IR
truthful menus
GAME THEORY
12
LECTURE 13
ADVANCED
EVOLUTION
Replicator
DYNAMICS
x dot
growth by payoff
GAME THEORY
13
LECTURE 14
ADVANCED
BARGAIN
Threat Point
SPLIT
NBS
outside option matters
GAME THEORY
14
LECTURE 15
ADVANCED
OLIGOPOLY
Reaction Curves
MARKET
q*
curves meet in output
GAME THEORY
15
LECTURE 16
ADVANCED
BEHAVIOR
Lab Evidence
EVIDENCE
lab
choice under bounds
GAME THEORY
16
LECTURE 17
ADVANCED
COMPUTE
Complexity
ALGORITHM
PPAD
search for equilibrium
GAME THEORY
17
LECTURE 18
APPLIED
MATCHING
Stable Allocation
MATCH
mu
stable pairs survive
GAME THEORY
18
LECTURE 19
APPLIED
NORMS
Coordination
FOCAL
risk
expectations coordinate
GAME THEORY
19
LECTURE 20
APPLIED
ENTRY
Deterrence
THREAT
commit
entry tests resolve
GAME THEORY
20
LECTURE 21
APPLIED
VOTING
Median Voter
SPACE
m
platforms chase median
GAME THEORY
21
LECTURE 22
APPLIED
COOPERATION
Prisoner's Dilemma
REPEAT
C/D
cooperation as design
GAME THEORY
22
Interactive Arcade
博弈小游戏厅
Cooperation
策略竞赛
Repeated Prisoner's Dilemma
参加多轮锦标赛，管理信誉、惩罚和合作阈值。
Matching
算法推演
Deferred Acceptance Lab
逐轮运行提议、暂留和拒绝，观察稳定匹配如何形成。
Nash
均衡调参
Cournot Reaction Arena
用产量滑杆对抗 AI 厂商，逼近最佳反应交点。
Auction
风险收益
Auction Bid Arena
在一价、二价和保留价环境中练习出价策略。
Dynamic
逆向归纳
Entry Deterrence Tree
选择承诺、进入和应战动作，检验威胁是否可信。
Voting
空间竞争
Median Voter Race
移动候选人、改变选民分布，争夺多数票。
Cooperation
搭便车
Public Goods Lab
调节贡献水平，与 AI 玩家共同决定公共池和个人收益。
Information
信念更新
Signaling Market
设计教育信号与雇主阈值，区分分离和混同均衡。
Norms
社会规范
Coordination Switch
在风险占优和收益占优之间推动群体跨过预期阈值。
Bargaining
分配谈判
Bargaining Table
设置报价、耐心和外部选项，判断协议能否达成。
Case Library
案例库
MIT 14.12 Handouts
Handouts 与 MIT PDF
Papers & Resources
论文、资源与英文工具
Route Map
学习路径
Six-week Game Theory Plan
从语言、均衡、时间、信念到应用的路线图
完成节点会高亮
01
Week 1
基础表示 · 建立语言
读 01-03，做完表示转换练习，把玩家、策略、收益、信息集写成统一模板。
能把任意故事改写成矩阵或博弈树。0/3
02
Week 2
静态均衡 · 识别均衡
读 04-06，先做支配剔除，再画最佳反应，最后用无差异条件解混合概率。
能判断纯/混合纳什均衡，并解释效率损失。0/3
03
Week 3
动态顺序 · 处理时间
读 07-09，从终端节点回推，区分可信威胁、SPNE 与重复互动惩罚。
能用逆向归纳拆动态承诺。0/3
04
Week 4
信息不对称 · 加入信念
读 10-12，练贝叶斯更新、信号成本、IC/IR 约束和菜单筛选。
能写出类型、信号、后验和激励相容条件。0/3
05
Week 5
进阶模型 · 扩展模型
读 13-17，把演化、议价、寡头、行为和计算统一为均衡概念的扩展。
能解释模型为什么从标准理性走向制度与算法。0/5
06
Week 6
应用场景 · 落地应用
读 18-22，用匹配、规范、进入、投票、合作五类案例做一次完整复盘。
能把现实案例拆成行动、信息、收益和均衡预测。0/5