Simulated Annealing OPLAB in NTUIM agenda Hill climbing

Simulated Annealing 報告者: 李怡緯 OPLAB in NTUIM

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

Hill Climbing ß ß ß In tree structure，DFS search。 Which node should be selected to expand? Greedy method。

ß Scheme of hill climbing Þ Þ Step 1: form a 1 -element stack consisting of root node. Step 2: test to see if the top element in the stack is a goal node. Step 3: remove the top element from the stack and expand the element. Add the descendants of the element into the stack ordered by the evaluation function Step 4: if the list is empty, then failure. Otherwise, go to Step 2.

8 -puzzle problem evaluation function f(n) = w(n) where w(n) is # of misplaced tiles in node n 2 3 1 5 1 4 8 6 8 7 7 Initial arrangement 2 3 4 6 5 Goal state of the puzzle

Simulated Annealing vs. Hill Climbing ß ß Hill Climbing select best in its level. local optimal. Simulated Annealing: random search neighbor node. Þ Þ If more appropriate than now, then taken from. According to the possibility to decide whether taken from.

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

Case研討: Traveling Salesman Problem ß Problem Definition: TSP is the problem of a salesman who wants to find, starting from his home town, a shortest possible trip through a given set of customer cities and to return to its home town; visiting exactly once each city.

ß ß 該如何使用SA處理之? random select 2 edges in a tour West Side East Side

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

Meta-heuristics 禁制搜尋法(Tabu Search, TS) ß 基因演算法(Genetic Algorithm, GA) ß 門檻接受法(Threshold Accepting, TA) ß 類神經網路(Neural Network, NN) ß 蟻群演算法(Ant Colony Optimization, ACO) ß

agenda ß ß ß Hill climbing V. S. Simulated Annealing(SA) SA歷史與簡介 SA之流程 SA研討: 如何處理TSP SA之目前應用 結論

Thanks for your listening