1 / 69

Why How We Learn Matters

Why How We Learn Matters. Russell Golman Scott E Page. Overview. BIG Picture Game Theory Basics Nash Equilibrium Equilibrium something about the other. Stability Basins Learning Rules Why Learning Matters. Big Question.

ira-hooper
Download Presentation

Why How We Learn Matters

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Why How We Learn Matters Russell Golman Scott E Page

  2. Overview • BIG Picture • Game Theory Basics • Nash Equilibrium • Equilibrium something about the other. • Stability • Basins • Learning Rules • Why Learning Matters

  3. Big Question If you had one word to describe the social, political, and economic worlds around you would you choose equilibrium or complex?

  4. Methodological Question Do we construct simple, illustrative, insight generating models (PD, Sandpile, El Farol) or do we construct high fidelity, realistic models?

  5. My Answer: BOTH! Both types of models are useful in their own right. In addition, each tells us something about the other.

  6. My Answer: BOTH! Knowledge of simple models helps us construct better high fidelity models. Large models show if insights from simple models still apply.

  7. Today’s Exercise How does how agents learn influence outcomes?

  8. Step Way Back Complex Adaptive System - Agents - Variation - Selection*

  9. Examples Best respond to current state Better respond Mimic best Mimic better Include portions of best or better Random with death of the unfit

  10. Equilibrium Science We can start by looking at the role that learning rules play in equilibrium systems. This will give us some insight into whether they’ll matter in complex systems.

  11. Game Theory • Players • Actions • Payoffs

  12. Players

  13. Actions Cooperate: C Defect: D

  14. Payoffs C D C D

  15. Best Responses C D C D

  16. Best Responses C D C D

  17. Nash Equilibrium C D C 2,2 D

  18. “Equilibrium” Based Science Step 1: Set up game Step 2: Solve for equilibrium Step 3: Show how equilibrium depends on parameters of model Step 4: Provide empirical support

  19. Is Equilibrium Enough? Existence: Equilibrium exists Stability: Equilibrium is stable Attainable: Equilibrium is attained by a learning rule.

  20. Stability Stability can only be defined relative to a learning dynamic. In dynamical systems, we often take that dynamic to be a best response function, but with human actors we need not assume people best respond.

  21. Existence Theorem Theorem: Finite number of players, finite set of actions, then there exists a Nash Equilibrium. Pf: Show best response functions are upper hemi continuous and then apply Kakutani’s fixed point theorem

  22. Battle of Sexes Game EF CG EF CG

  23. Three Equilibria 1/4 3/4 3/4 1/4

  24. Unstable Mixed? 1/4 +e 3/4 - e EF 3/4 + 3e 3/4 - 3e CG

  25. Note the Implicit Assumption Our stability analysis assumed that Player 1 would best respond to Player 2’s tremble. However, the learning rule could be go to the mixed strategy equilibrium. If so, Player 1 would sit tight and Player 2 would return to the mixed strategy equilibrium.

  26. Empirical Foundations We need to have some understanding of how people learn and adapt to say anything about stability.

  27. Classes of Learning Rules Belief Based Learning Rules: People best respond given their beliefs about how other people play. Replicator Learning Rules: People replicate successful actions of others.

  28. Examples Belief Based Learning Rules: Best response functions Replicator Learning Rules: Replicator dynamics

  29. Stability Results An extensive literature provides conditions (fairly week) under which the two learning rules have identical stability property. Synopsis: Learning rules do not matter

  30. Basins Question Do games exist in which best response dynamics and replicator dynamics produce very different basins of attraction? Question: Does learning matter?

  31. Best Response Dynamics x = mixed strategy of Player 1 y = mixed strategy of Player 2 dx/dt = BR(y) - x dy/dt = BR(x) - y

  32. Replicator Dynamics dxi/dt = xi( i - ave) dyi/dt = yi( i - ave)

  33. Symmetric Matrix Game A B C A B C

  34. Conceptualization Imagine a large population of agents playing this game. Each chooses an action. The population distribution over actions creates a mixed strategy. We can then study the dynamics of that population given different learning rules.

  35. Best Response Basins A B C A B C A > B iff 60pA + 60pB + 30pC > 30pA + 70pB + 20pC A > C iff 60pA + 60pB + 30pC > 70pA + 25pB + 35pC

  36. Best Response Basins C C A C B B B A

  37. Stable Equilibria C C A C B B B A

  38. Best Response Basins C C A A B B B A

  39. Replicator Dynamics C C A ? B B B A

  40. Replicator Dynamics Basins C C A A B B B B A

  41. Recall: Basins Question Do games exist in which best response dynamics and replicator dynamics produce very different basins of attraction? Question: Does learning matter?

  42. Conjecture For any  > 0, There exists a symmetric matrix game such that the basins of attraction for distinct equilibria under continuous time best response dynamics and replicator dynamics overlap by less than 

  43. Results Thm 1 (SP): Can be done if number of actions goes to infinity

  44. Results Thm 1 (SP): Can be done if number of actions goes to infinity Thm 2 (RG): Can be done if number of actions scales with 1/

  45. Results Thm 1 (SP): Can be done if number of actions goes to infinity Thm 2 (RG): Can be done if number of actions scales with 1/ Thm 3 (RG): Cannot be done with two actions.

  46. Results Thm 1 (SP): Can be done if number of actions goes to infinity Thm 2 (RG): Can be done if number of actions scales with 1/ Thm 3 (RG): Cannot be done with two actions. Thm 4 (SP): Can be done with four!

  47. Collective Action Game SI Coop Pred Naive SI Coop Pred Naive

  48. Intuition: Naïve Goes Away Pred Coop SI

  49. Intuition: Naïve Goes Away Pred SI Coop Coop SI

More Related