---

tags:

• 🌱
• AI
• ComputerScience date: 18--Apr--2023

---

No player can unilaterally change strategy to receive better reward while assuming other players stay with their strategy choice

Every player chooses the best response against their opponent. If in Nash Equilibrium then no player can perform better by changing strategy.

• Each player in game can have one or more local stable point
  • This is known as a strategy set → When there are multiple Nash Equilibria (strategy)
• Each stable point is a Nash Equilibria
  • A Dominant strategy is the best response to all action and thus any Dominant strategy equilibrium must be a Nash Equilibrium

• Prisoner Dilemma case, the convergence is when both confesses which is worse off than both staying silent
  • This is because each player performs with maximum self benefit in mind (selfish plays)

---

Links: