duality ii
play

Duality (II) Lijun Zhang zlj@nju.edu.cn http://cs.nju.edu.cn/zlj - PowerPoint PPT Presentation

Duality (II) Lijun Zhang zlj@nju.edu.cn http://cs.nju.edu.cn/zlj Outline Saddle-point Interpretation Max-min Characterization of Weak and Strong Duality Saddle-point Interpretation Game Interpretation Optimality


  1. Duality (II) Lijun Zhang zlj@nju.edu.cn http://cs.nju.edu.cn/zlj

  2. Outline  Saddle-point Interpretation  Max-min Characterization of Weak and Strong Duality  Saddle-point Interpretation  Game Interpretation  Optimality Conditions  Certificate of Suboptimality and Stopping Criteria  Complementary Slackness  KKT Optimality Conditions  Solving the Primal Problem via the Dual  Examples  Generalized Inequalities

  3. Outline  Saddle-point Interpretation  Max-min Characterization of Weak and Strong Duality  Saddle-point Interpretation  Game Interpretation  Optimality Conditions  Certificate of Suboptimality and Stopping Criteria  Complementary Slackness  KKT Optimality Conditions  Solving the Primal Problem via the Dual  Examples  Generalized Inequalities

  4. More Symmetric Form  Assume no equality constraint � sup 𝑀 𝑦, 𝜇 � sup 𝑔 � 𝑦 � � 𝜇 � 𝑔 � 𝑦 �≽� �≽� ��� � �𝑔 � 𝑦 𝑔 � 𝑦 � 0, 𝑗 � 1, … , 𝑛 ∞ otherwise  Suppose � for some . Then, by � and �≽� �  If � , then the optimal choice of is and �≽� �

  5. More Symmetric Form  Optimal Value of Primal Problem 𝑞 ⋆ � inf � sup 𝑀 𝑦, 𝜇 �≽�  Optimal Value of Dual Problem 𝑒 ⋆ � sup inf � 𝑀 𝑦, 𝜇 �≽�  Weak Duality sup inf � 𝑀 𝑦, 𝜇 � inf � sup 𝑀 𝑦, 𝜇 �≽� �≽�  Strong Duality sup inf � 𝑀 𝑦, 𝜇 � inf � sup 𝑀 𝑦, 𝜇 �≽� �≽�  Min and Max can be switched

  6. A More General Form  Max-min Inequality sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈� � �  For any and any � �  Strong Max-min Property sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈�  Hold only in special cases

  7. Outline  Saddle-point Interpretation  Max-min Characterization of Weak and Strong Duality  Saddle-point Interpretation  Game Interpretation  Optimality Conditions  Certificate of Suboptimality and Stopping Criteria  Complementary Slackness  KKT Optimality Conditions  Solving the Primal Problem via the Dual  Examples  Generalized Inequalities

  8. Saddle-point Interpretation  is a saddle point for 𝑔 𝑥 �, 𝑨 � 𝑔 𝑥 �, 𝑨̃ � 𝑔 𝑥, 𝑨̃ , ∀𝑥 ∈ 𝑋, 𝑨 ∈ 𝑎  minimizes , maximizes 𝑔 𝑥 �, 𝑨̃ � inf �∈� 𝑔�𝑥, 𝑨̃� , 𝑔 𝑥 �, 𝑨̃ � sup 𝑔 𝑥 �, 𝑨 �∈� https: / / en.wikipedia.org/ wiki/ Saddle_point

  9. Saddle-point Interpretation  is a saddle point for 𝑔 𝑥 �, 𝑨 � 𝑔 𝑥 �, 𝑨̃ � 𝑔 𝑥, 𝑨̃ , ∀𝑥 � ∈ 𝑋, 𝑨̃ ∈ 𝑎  minimizes , maximizes 𝑔 𝑥 �, 𝑨̃ � inf �∈� 𝑔�𝑥, 𝑨̃� , 𝑔 𝑥 �, 𝑨̃ � sup 𝑔 𝑥 �, 𝑨 �∈�  Imply the strong max-min property sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� 𝑔�𝑥, 𝑨̃� � 𝑔 𝑥 �, 𝑨̃ �∈� 𝑔 𝑥, 𝑨 � 𝑔 𝑥 �, 𝑨̃ � sup 𝑔 𝑥 �, 𝑨 � inf �∈� sup �∈� �∈� ⇒ sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈� ⇒ sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈�

  10. Saddle-point Interpretation  is a saddle point for 𝑔 𝑥 �, 𝑨 � 𝑔 𝑥 �, 𝑨̃ � 𝑔 𝑥, 𝑨̃ , ∀𝑥 � ∈ 𝑋, 𝑨̃ ∈ 𝑎  minimizes , maximizes 𝑔 𝑥 �, 𝑨̃ � inf �∈� 𝑔�𝑥, 𝑨̃� , 𝑔 𝑥 �, 𝑨̃ � sup 𝑔 𝑥 �, 𝑨 �∈� ⋆ are primal and dual optimal ⋆  If ⋆ ⋆ points and strong duality holds, form a saddle-point.  If is saddle-point, then is primal optimal, is dual optimal, and the duality gap is zero.

  11. Outline  Saddle-point Interpretation  Max-min Characterization of Weak and Strong Duality  Saddle-point Interpretation  Game Interpretation  Optimality Conditions  Certificate of Suboptimality and Stopping Criteria  Complementary Slackness  KKT Optimality Conditions  Solving the Primal Problem via the Dual  Examples  Generalized Inequalities

  12. Continuous Zero-sum Game  Two players  The 1st player chooses , and the 2nd player selects  Player 1 pays an amount to player 2  Goals  Player 1 wants to minimize  Player 2 wants to maximize  Continuous game  The choices are vectors, and not discrete

  13. Continuous Zero-sum Game  Player 1 makes his choice first  Player 2 wants to maximize payoff and the resulting payoff is �∈�  Player 1 knows that player 2 will follow this strategy, and so will choose to make as small as possible �∈�  Thus, player 1 chooses argmin sup 𝑔 𝑥, 𝑨 �∈� �∈�  The payoff �∈� sup inf 𝑔 𝑥, 𝑨 �∈�

  14. Continuous Zero-sum Game  Player 2 makes his choice first  Player 1 wants to minimize payoff and the resulting payoff is �∈�  Player 2 knows that player 1 will follow this strategy, and so will choose to make as large as possible �∈�  Thus, player 2 chooses argmax �∈� 𝑔 𝑥, 𝑨 inf �∈�  The payoff sup �∈� 𝑔 𝑥, 𝑨 inf �∈�

  15. Continuous Zero-sum Game  Max-min Inequality sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈� Player 2 plays first Player 1 plays first  Player 1 wants to minimize  Player 2 wants to maximize It is better for a player to go second

  16. Continuous Zero-sum Game  Strong Max-min Property sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈� Player 2 plays first Player 1 plays first  Player 1 wants to minimize  Player 2 wants to maximize There is no advantage to playing second

  17. Continuous Zero-sum Game  Strong Max-min Property sup �∈� 𝑔 𝑥, 𝑨 � inf inf �∈� sup 𝑔 𝑥, 𝑨 �∈� �∈� Player 2 plays first Player 1 plays first  Saddle-point Property  If is a saddle-point for (and ), then it is called a solution of the game  𝑥 � : the optimal strategy for player 1  𝑨̃ : the optimal strategy for player 2  No advantage to playing second

  18. A Special Case � �  Payoff is the Lagrangian; �  Player 1 chooses the primal variable while player 2 chooses the dual variable  The optimal choice for player 2, if she must choose first, is any dual optimal ⋆  The resulting payoff: 𝑒 ⋆  Conversely, if player 1 chooses first, his ⋆ optimal choice is any primal optimal  The resulting payoff: 𝑞 ⋆  Duality gap: advantage of going second

  19. Outline  Saddle-point Interpretation  Max-min Characterization of Weak and Strong Duality  Saddle-point Interpretation  Game Interpretation  Optimality Conditions  Certificate of Suboptimality and Stopping Criteria  Complementary Slackness  KKT Optimality Conditions  Solving the Primal Problem via the Dual  Examples  Generalized Inequalities

  20. Certificate of Suboptimality  Dual Feasible  A lower bound on the optimal value of the primal problem ⋆  Provides a proof or certificate  Bound how suboptimal a given feasible ⋆ point is, without knowing the value of ⋆ � �  𝑦 is 𝜗 -suboptimal for primal problem  ( 𝜇, 𝜉� is 𝜗 -suboptimal for dual

  21. Certificate of Suboptimality  Gap between Primal & Dual Objectives �  Referred to as duality gap associated with primal feasible and dual feasible  localizes the optimal value of the primal (and dual) problems to an interval ⋆ ⋆ � �  The width of the interval is the duality gap  If duality gap of is , then is primal optimal and is dual optimal

  22. Stopping Criteria  Optimization algorithms produce a � and dual sequence of primal feasible � � feasible for  Required absolute accuracy: ���  A Nonheuristic Stopping Criterion � � � � ��� �  Guarantees when algorithm terminates, is ��� -suboptimal

  23. Stopping Criteria  A Relative Accuracy ���  Nonheuristic Stopping Criteria  If � 𝑕 𝜇 � , 𝜉 � � 𝑦 � 𝑔 𝑕 𝜇 � , 𝜉 � � 0, � 𝜗 ��� 𝑕 𝜇 � , 𝜉 � or � 𝑕 𝜇 � , 𝜉 � � 𝑦 � 𝑔 � 𝑦 � 𝑔 � 0, � 𝜗 ��� � 𝑦 � �𝑔 ⋆  Then , and the relative error satisfies � 𝑦 � � 𝑞 ⋆ 𝑔 � 𝜗 ��� |𝑞 ⋆ |

  24. Outline  Saddle-point Interpretation  Max-min Characterization of Weak and Strong Duality  Saddle-point Interpretation  Game Interpretation  Optimality Conditions  Certificate of Suboptimality and Stopping Criteria  Complementary Slackness  KKT Optimality Conditions  Solving the Primal Problem via the Dual  Examples  Generalized Inequalities

Recommend


More recommend