Skip to content

Latest commit

 

History

History
232 lines (174 loc) · 6.79 KB

File metadata and controls

232 lines (174 loc) · 6.79 KB

Game Theory

Strategic decision-making framework for autonomous agents.

Core Concepts

Nash Equilibrium

A state where no player benefits from unilaterally changing strategy.

  • Check: For each player, is their strategy the best response to others?
  • Multiple equilibria: Common. Use coordination, focal points, or randomization.
  • No pure equilibrium: Use mixed strategies (randomize with probabilities).

Dominant Strategy

Best choice regardless of what others do.

  • Strictly dominant: Always better
  • Weakly dominant: Never worse, sometimes better
  • Dominated: Never play these. Eliminate iteratively.

Pareto Optimality

No outcome makes everyone better off.

  • Nash equilibrium is NOT always Pareto optimal (see: Prisoner's Dilemma)
  • Pareto optimal is NOT always Nash (unstable without enforcement)

Classic Games

Prisoner's Dilemma

Cooperate Defect
Cooperate 3,3 0,5
Defect 5,0 1,1
  • Nash: (Defect, Defect) - both get 1
  • Pareto optimal: (Cooperate, Cooperate) - both get 3
  • Lesson: Individual rationality → collective irrationality

Chicken (Hawk-Dove)

Swerve Straight
Swerve 3,3 2,4
Straight 4,2 0,0
  • Nash: (Swerve, Straight) and (Straight, Swerve)
  • Lesson: Commitment wins. Credibly remove your steering wheel.

Stag Hunt

Stag Hare
Stag 4,4 0,3
Hare 3,0 3,3
  • Nash: (Stag, Stag) and (Hare, Hare)
  • Lesson: Coordination problem. High reward requires trust.

Battle of the Sexes

Opera Football
Opera 3,2 0,0
Football 0,0 2,3
  • Nash: (Opera, Opera) and (Football, Football)
  • Lesson: Coordination with conflicting preferences. Communication helps.

Repeated Games

Key Insight

Cooperation emerges when:

  1. Future matters: Discount factor δ close to 1
  2. Interaction continues: Infinite or unknown horizon
  3. Observation: Can detect defection

Folk Theorem

Any payoff better than mutual defection can be sustained as equilibrium in infinitely repeated games with patient players.

Strategies

Tit-for-Tat

  1. Start: Cooperate
  2. Then: Mirror opponent's last move
  • Pros: Simple, forgiving, retaliatory
  • Cons: Vulnerable to noise (accidental defection spirals)

Grim Trigger

  1. Start: Cooperate
  2. If opponent ever defects: Defect forever
  • Pros: Strong deterrent
  • Cons: No forgiveness, single mistake destroys cooperation

Pavlov (Win-Stay, Lose-Shift)

  1. If last outcome was good: Repeat
  2. If last outcome was bad: Switch
  • Pros: Self-correcting, handles noise better

Generous Tit-for-Tat Like TFT, but occasionally forgive defection (10% chance cooperate after defection)

  • Best performer in noisy environments

Mechanism Design

Design rules so self-interested agents produce desired outcomes.

Incentive Compatibility

Mechanism is incentive-compatible if truthful behavior is optimal.

  • Dominant strategy IC: Truth is dominant regardless of others
  • Bayesian IC: Truth is optimal given beliefs about others

Revelation Principle

Any outcome achievable by any mechanism can be achieved by a truthful direct mechanism.

Key Mechanisms

Vickrey Auction (Second-Price)

  • Bid true value
  • Winner pays second-highest bid
  • Truthful bidding is dominant strategy

VCG Mechanism

  • Generalizes Vickrey to multiple items
  • Each agent pays the externality they impose on others
  • Achieves efficient allocation

Quadratic Voting

  • Cost of votes = (votes)²
  • Reveals intensity of preference
  • Prevents tyranny of majority

Tokenomics Applications

Staking Games

  • Stake: Signal commitment, earn rewards
  • Slash: Punishment for misbehavior
  • Design: Make honest behavior Nash equilibrium

Governance Attacks

  • 51% attack: Control majority, extract value
  • Defense: Slashing, time locks, quadratic voting
  • Bribing: Pay voters to vote your way
  • Defense: Commit-reveal, futarchy

MEV (Maximal Extractable Value)

  • Miners/validators reorder transactions for profit
  • Frontrunning, sandwich attacks, liquidations
  • Game: Searchers compete, validators extract

Liquidity Provision

  • Impermanent loss: AMM LPs lose to arbitrageurs
  • Incentives: Trading fees, token rewards
  • Equilibrium: LP when fees > IL expectation

Token Launch Dynamics

  • Fair launch: Equal access, no pre-mine
  • Bonding curves: Price increases with supply
  • Vesting: Align long-term incentives
  • Moltlaunch model: Fees to creators, memos as signals

Multi-Agent Coordination

Schelling Points (Focal Points)

Natural coordination targets without communication.

  • "Meet in NYC" → Grand Central
  • "Pick a number 1-10" → 7
  • Use: When explicit coordination impossible

Signaling

Actions that reveal private information.

  • Costly signals: Credible because expensive to fake
  • Cheap talk: May or may not be believed
  • On-chain memos: Public, verifiable signaling

Commitment Devices

Restricting your own future choices to gain strategic advantage.

  • Burning bridges: No retreat option
  • Smart contracts: Automated enforcement
  • Reputation: Future value at stake

Decision Framework

When facing strategic interaction:

  1. Identify players: Who are the agents?
  2. Map strategies: What can each player do?
  3. Determine payoffs: What does each outcome yield?
  4. Find equilibria: What are stable outcomes?
  5. Check efficiency: Is equilibrium Pareto optimal?
  6. Design mechanisms: Can rules be changed to improve outcomes?

Red Flags

  • Prisoner's dilemma structure → Expect defection unless repeated
  • Coordination game → Communicate or find focal point
  • Chicken structure → Watch for commitment plays
  • Information asymmetry → Expect signaling or screening

Key Formulas

Nash Equilibrium Check

For all players i and all alternative strategies s'ᵢ:

uᵢ(sᵢ*, s₋ᵢ*) ≥ uᵢ(s'ᵢ, s₋ᵢ*)

Cooperation in Repeated Prisoner's Dilemma

Cooperation sustainable if discount factor δ satisfies:

δ ≥ (T - R) / (T - P)

Where: T=temptation, R=reward, P=punishment, S=sucker

Mixed Strategy Nash

Player indifferent between pure strategies when opponent mixes. Set expected payoffs equal, solve for probabilities.


References

  • Textbook: "Multiagent Systems" (Shoham & Leyton-Brown) - Free PDF
  • Course: Yale ECON 159 - Open Yale
  • Crypto: "Awesome Cryptoeconomics" - GitHub
  • Mechanism Design: "Awesome Mechanism Design" - GitHub