multi agent learning
play

Multi-agent learning Satising pla y Gerard Vreeswijk , Intelligent - PowerPoint PPT Presentation

Jul 08, 2023 •406 likes •2.23k views

Multi-agent learning Satising pla y Gerard Vreeswijk , Intelligent Software Systems, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands. Tuesday 16 th June, 2020 Assumptions in game playing Author: Gerard

  1. What if none of these assumptions holds? ■ Players don’t know the This takes the problem out of structure of the game. game theory into machine learning. ● Players don’t know who’s What can we do? playing. ■ Reinforcement learning. ● Players don’t know the arsenal of other players. Disadvantages: ■ Players can’t observe other ● No reference to past player’s actions. average payoffs. ■ Players can’t observe other ● Difficult theory. player’s payoffs. Alternative: ■ Players aren’t aware that they ■ Satisficing learning. are in a game. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 3

  2. Herbert A. Simon on maximising vs. satisficing Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  3. Herbert A. Simon on maximising vs. satisficing Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  4. Herbert A. Simon on maximising vs. satisficing Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  5. Herbert A. Simon on maximising vs. satisficing “A decision maker who chooses the best available alternative ac- cording to some criteria is said to optimise; one who chooses an alternative that meets or exceeds specified criteria, but that is not guaranteed to be either unique or in any sense the best, is said to satisfice .” a a H. Simon “Models of bounded rationality” in: Empirically grounded economic reasons , Vol. 3. MIT Press, 1997. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  6. Herbert A. Simon on maximising vs. satisficing “A decision maker who chooses the best available alternative ac- cording to some criteria is said to optimise; one who chooses an alternative that meets or exceeds specified criteria, but that is not guaranteed to be either unique or in any sense the best, is said to satisfice .” a a H. Simon “Models of bounded rationality” in: Empirically grounded economic reasons , Vol. 3. MIT Press, 1997. “Decision makers can satisfice either by finding optimum solu- tions for a simplified world, or by finding satisfactory solutions for a more realistic world. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  7. Herbert A. Simon on maximising vs. satisficing “A decision maker who chooses the best available alternative ac- cording to some criteria is said to optimise; one who chooses an alternative that meets or exceeds specified criteria, but that is not guaranteed to be either unique or in any sense the best, is said to satisfice .” a a H. Simon “Models of bounded rationality” in: Empirically grounded economic reasons , Vol. 3. MIT Press, 1997. “Decision makers can satisfice either by finding optimum solu- tions for a simplified world, or by finding satisfactory solutions for a more realistic world. Neither approach, in general, domi- nates the other, and both have continued to co-exist in the world of management science.” Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  8. Herbert A. Simon on maximising vs. satisficing “A decision maker who chooses the best available alternative ac- cording to some criteria is said to optimise; one who chooses an alternative that meets or exceeds specified criteria, but that is not guaranteed to be either unique or in any sense the best, is said to satisfice .” a a H. Simon “Models of bounded rationality” in: Empirically grounded economic reasons , Vol. 3. MIT Press, 1997. “Decision makers can satisfice either by finding optimum solu- tions for a simplified world, or by finding satisfactory solutions for a more realistic world. Neither approach, in general, domi- nates the other, and both have continued to co-exist in the world of management science.” a a H. Simon “Rational decision making in business organizations” in: The American Economic Review, Vol. 69(4), pp. 493-513. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 4

  9. Karandikar et al. ’s algorithm for satisficing play (1989) Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 5

  10. aspiration level p ersisten e rate Satisficing algorithm Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  11. aspiration level p ersisten e rate Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  12. aspiration level p ersisten e rate Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  13. p ersisten e rate Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . aspiration Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  14. p ersisten e rate Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . Updated as aspiration α t + 1 = Def λα t + ( 1 − λ ) π t , Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  15. Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . Updated as aspiration α t + 1 = Def λα t + ( 1 − λ ) π t , where λ is the p ersisten e rate Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  16. Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . Updated as aspiration α t + 1 = Def λα t + ( 1 − λ ) π t , where λ is the rate , and π t is the payoff in round t . p ersisten e Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  17. Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . Updated as aspiration α t + 1 = Def λα t + ( 1 − λ ) π t , where λ is the rate , and π t is the payoff in round t . p ersisten e ● It is up to the programmer to choose an initial action A 0 , and an initial aspiration level α 0 . Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  18. Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . Updated as aspiration α t + 1 = Def λα t + ( 1 − λ ) π t , where λ is the rate , and π t is the payoff in round t . p ersisten e ● It is up to the programmer to choose an initial action A 0 , and an initial aspiration level α 0 . ■ Satisficing algorithm: � A t if π t ≥ α t , A t + 1 = any other action else. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  19. Satisficing algorithm ■ At any time, t , the agent’s state is a tuple ( A t , α t ) . A t is the current action. ● ● α t is the current level . Updated as aspiration α t + 1 = Def λα t + ( 1 − λ ) π t , where λ is the rate , and π t is the payoff in round t . p ersisten e ● It is up to the programmer to choose an initial action A 0 , and an initial aspiration level α 0 . ■ Satisficing algorithm: � A t if π t ≥ α t , A t + 1 = any other action else. Also works if “any other action” is replaced by “any action”. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 6

  20. Example of satisficing play Game: prisoner’s dilemma. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  21. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  22. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  23. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  24. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  25. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  26. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  27. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  28. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  29. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  30. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  31. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  32. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  33. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  34. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  35. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  36. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  37. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  38. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  39. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  40. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  41. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  42. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  43. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  44. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  45. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  46. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  47. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  48. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  49. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  50. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  51. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  52. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  53. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  54. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  55. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  56. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  57. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  58. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  59. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  60. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  61. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  62. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  63. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  64. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  65. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  66. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  67. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  68. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  69. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  70. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  71. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  72. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C 0 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  73. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C 0 2.01171875 Progress of aspirations. Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  74. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C 0 2.01171875 Progress of aspirations. 10 Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  75. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C 0 2.01171875 Progress of aspirations. 10 C Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  76. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C 0 2.01171875 Progress of aspirations. 10 C D Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

  77. Example of satisficing play Game: prisoner’s dilemma. Strategy player 1: tit-for-tat. Strategy player 2: satisficing with initial state ( A 0 , α 0 ) = ( C , 5 ) . Persistence rate: λ = 0.5. t TFT A t π t α t 0 C C 3 5 Α 5 1 C D 5 4 4 2 D D 1 4.5 3 D C 0 2.75 3 4 C D 5 1.375 5 D D 1 3.1875 2 6 D C 0 2.09375 1 7 C D 5 1.046875 8 D D 1 3.0234375 t 0 2 4 6 8 10 9 D C 0 2.01171875 Progress of aspirations. 10 C D 5 Author: Gerard Vreeswijk. Slides last modified on June 16 th , 2020 at 12:21 Multi-agent learning: Satisficing play, slide 7

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Multi-agent learning Multi-agent reinforcement learning Gerard Vreeswijk , Intelligent Systems

Multi-agent learning Multi-agent reinforcement learning Gerard Vreeswijk , Intelligent Systems

Multi-agent learning Multi-agent reinforcement learning Multi-agent learning Multi-agent reinforcement learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands.

752 views • 21 slides
Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department,

Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department,

Multi-agent learning Multi-agent reinforcement learning Multi-agent reinfo rement lea rning Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The

528 views • 21 slides
REINFORCEMENT LEARNING IN MULTI-AGENT SYSTEMS MACHINE LEARNING MEETUP DR. ANA PELETEIRO

REINFORCEMENT LEARNING IN MULTI-AGENT SYSTEMS MACHINE LEARNING MEETUP DR. ANA PELETEIRO

REINFORCEMENT LEARNING IN MULTI-AGENT SYSTEMS MACHINE LEARNING MEETUP DR. ANA PELETEIRO RAMALLO 29-08-2016 TABLE OF CONTENTS MULTI-AGENT SYSTEMS GAME THEORY REINFORCEMENT LEARNING MULTI-AGENT LEARNING 2 ZALANDO Our purpose: to Zalando

1.45k views • 20 slides
ROMA: Multi-Agent Reinforcement Learning with Emerging Roles Tonghan Wang, Heng Dong, Victor

ROMA: Multi-Agent Reinforcement Learning with Emerging Roles Tonghan Wang, Heng Dong, Victor

ROMA: Multi-Agent Reinforcement Learning with Emerging Roles Tonghan Wang, Heng Dong, Victor Lesser, Chongjie Zhang Tsinghua University, UMass Amherst Multi-Agent Systems Robot Football Game Multi-Agent Assembly One Major Challenge of

426 views • 30 slides
Overview Multi-Agent Systems Introduction to multi-agent systems and agent societies Agent

Overview Multi-Agent Systems Introduction to multi-agent systems and agent societies Agent

CPE/CSC 580-S06 Artificial Intelligence Intelligent Agents Overview Multi-Agent Systems Introduction to multi-agent systems and agent societies Agent Communication knowledge exchange among agents Agent Interaction eliminates explicit

623 views • 26 slides
QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement

QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement

QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement Learning Kyunghwan Son , Daewoo Kim, Wan Ju Kang, David Hostallero, Yung Yi School of Electrical Engineering, KAIST Cooperative Multi-Agent Reinforcement

662 views • 8 slides
Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department,

Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department,

Multi-agent learning Fictitious Play Fititious Pla y Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands. Last modified on February 27 th ,

602 views • 41 slides
YunQi 2050 - DRL Session Communication in Multi-agent Reinforcement Learning Ying Wen

YunQi 2050 - DRL Session Communication in Multi-agent Reinforcement Learning Ying Wen

YunQi 2050 - DRL Session Communication in Multi-agent Reinforcement Learning Ying Wen Department of Computer Science, University College London MediaGamma Ltd. ying.wen@cs.ucl.ac.uk 30 May, 2018 Multi-agent in Real-World Human

402 views • 27 slides
Multi-agent learning Emergence of Conventions Gerard Vreeswijk , Intelligent Systems Group,

Multi-agent learning Emergence of Conventions Gerard Vreeswijk , Intelligent Systems Group,

Multi-agent learning Emergence of Conventions Multi-agent learning Emergence of Conventions Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands. Last modified on

742 views • 48 slides
W HAT S AN A GENT ? Weiss, p. 29 [after Wooldridge and Jennings]: An agent is a

W HAT S AN A GENT ? Weiss, p. 29 [after Wooldridge and Jennings]: An agent is a

Todays Class M ULTI -A GENT S YSTEMS Overview and Research Directions Whats an agent? AI Class 12 (C H . 17.517.6) Multi-Agent Systems Cooperative multi-agent systems Competitive multi-agent systems Game time! MAS

309 views • 7 slides
Social Learning in Multi Agent Multi Armed Bandits Abishek Sankararaman, UC Berkeley April 9,

Social Learning in Multi Agent Multi Armed Bandits Abishek Sankararaman, UC Berkeley April 9,

Social Learning in Multi Agent Multi Armed Bandits Abishek Sankararaman, UC Berkeley April 9, 2020 Joint Work with - Sanjay Shakkottai, Ronshee Chawla, UT Austin - Ayalvadi Ganesh, University of Bristol Multi Armed Bandit Problem A set of

973 views • 54 slides
Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department,

Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department,

Multi-agent learning Gradient Dynamics Gradient Dynamis Multi-agent learning Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands. Last modified on March 1 st ,

394 views • 28 slides
Multi-agent learning Repeated games Gerard Vreeswijk , Intelligent Systems Group, Computer Science

Multi-agent learning Repeated games Gerard Vreeswijk , Intelligent Systems Group, Computer Science

Multi-agent learning Repeated games Multi-agent learning Repeated games Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands. Last modified on February 9 th , 2012 at

1.33k views • 36 slides
Multi-agent learning Erik Berbee & Bas van Gijzel , Master Student AT, Utrecht University Erik

Multi-agent learning Erik Berbee & Bas van Gijzel , Master Student AT, Utrecht University Erik

Multi-agent learning Methodology of MAL research Metho dology of MAL resea rh Multi-agent learning Erik Berbee & Bas van Gijzel , Master Student AT, Utrecht University Erik Berbee & Bas van Gijzel. Slides last processed on Monday

1.49k views • 35 slides
SIKS tutorial Agent Systems Multi-agent learning Gerard Vreeswijk Intelligent Systems

SIKS tutorial Agent Systems Multi-agent learning Gerard Vreeswijk Intelligent Systems

Introduction Cournot dynamics Alternative Cournot dynamics SIKS tutorial Agent Systems Multi-agent learning Gerard Vreeswijk Intelligent Systems Group, Computer Science Department Faculty of Sciences, Utrecht University, The Netherlands

1.45k views • 141 slides
Multi-agent learning Fictitious Play Gerard Vreeswijk , Intelligent Systems Group, Computer

Multi-agent learning Fictitious Play Gerard Vreeswijk , Intelligent Systems Group, Computer

Multi-agent learning Fictitious Play Multi-agent learning Fictitious Play Gerard Vreeswijk , Intelligent Systems Group, Computer Science Department, Faculty of Sciences, Utrecht University, The Netherlands. Gerard Vreeswijk. Slides last

650 views • 39 slides
WELCOME! DTRF Western Regional Patient Meeting South San Francisco, April 13, 2019 Our Mission

WELCOME! DTRF Western Regional Patient Meeting South San Francisco, April 13, 2019 Our Mission

WELCOME! DTRF Western Regional Patient Meeting South San Francisco, April 13, 2019 Our Mission to aggressively fund research to accelerate the development of improved therapies, and ultimately find a cure for desmoid tumors. We collaborate

377 views • 16 slides
Protocol to Patient (P2P) Ghulam Warsi 1 , Kert Viele 2 , Lebedinsky Claudia 1 , , Parasuraman

Protocol to Patient (P2P) Ghulam Warsi 1 , Kert Viele 2 , Lebedinsky Claudia 1 , , Parasuraman

Protocol to Patient (P2P) Ghulam Warsi 1 , Kert Viele 2 , Lebedinsky Claudia 1 , , Parasuraman Sudha 1 , Eric Slosberg 1 , Barinder Kang 1 , August Salvado 1 , Lening Zhang 1 , Donald A. Berry 2 1 Novartis Pharmaceuticals Corporation, East Hanover,

620 views • 20 slides
Cancer treatment in Mexico and the Seguro Popular August, 2012 1 Contents 1. Cancer as a public

Cancer treatment in Mexico and the Seguro Popular August, 2012 1 Contents 1. Cancer as a public

Cancer treatment in Mexico and the Seguro Popular August, 2012 1 Contents 1. Cancer as a public health and economic priority issue 2. Mexicos Social Protection in Health System 3. Analysis of cancer treatments covered by the Catastrophic

484 views • 23 slides
Rx: Medical Neighborhoods The Medical Legal Partnership and Social Determinants of Health

Rx: Medical Neighborhoods The Medical Legal Partnership and Social Determinants of Health

Rx: Medical Neighborhoods The Medical Legal Partnership and Social Determinants of Health Carolyn Pointer, JD April 25, 2017 FINANCIAL DISCLOSURE Neither Carolyn Pointer nor her husband have a personal or professional financial relationship

579 views • 43 slides
Bounded rationality in the description of research excellence in Early Career Researchers in the

Bounded rationality in the description of research excellence in Early Career Researchers in the

Bounded rationality in the description of research excellence in Early Career Researchers in the Social Sciences and Humanities Authors: Derrick, G., de Jong, S., Muhonen, R., Afxentiou G., Urbanc M. ECRs Evaluation System Introduction

238 views • 8 slides
On Satisficing Planning with Admissible Empirical Evaluation Heuristics Discussion Summary and

On Satisficing Planning with Admissible Empirical Evaluation Heuristics Discussion Summary and

Background Dominating Actions On Satisficing Planning with Admissible Empirical Evaluation Heuristics Discussion Summary and Future Work R. Bahumi C. Domshlak M. Katz Faculty of Industrial Engineering and Management Technion - Israel

500 views • 17 slides
Know ledge-Based Systems IS430 Decision Making, Systems, Modeling, and Support Mostafa Z. Ali

Know ledge-Based Systems IS430 Decision Making, Systems, Modeling, and Support Mostafa Z. Ali

Winter 2009 Lecture 2 Know ledge-Based Systems IS430 Decision Making, Systems, Modeling, and Support Mostafa Z. Ali Mostafa Z. Ali mzali@just.edu.jo Lecture 2: Slide 1 Decision Making: Introduction and Definitions Characteristics of

778 views • 42 slides
An Overview of the International Planning Competition Part 1: Classical Tracks Amanda Coles 1

An Overview of the International Planning Competition Part 1: Classical Tracks Amanda Coles 1

AI Planning and the IPC Classical Tracks Get Involved An Overview of the International Planning Competition Part 1: Classical Tracks Amanda Coles 1 Andrew Coles 1 Alvaro Torralba 2 Florian Pommerening 3 1 Kings College London 2 Saarland

1.68k views • 138 slides

More recommend