Skip to content

Latest commit

 

History

History
34 lines (27 loc) · 1.72 KB

review3.md

File metadata and controls

34 lines (27 loc) · 1.72 KB

home | copyright ©2016, [email protected]

overview | syllabus | src | submit | chat

Review-3

Lecture Date : 09/06/16

Theory

  1. What is ordered and unordered search. In what way are they different?
  2. What is the difference between a local and global maxima?
  3. What is the drawback of greedy search?
  4. Why do you think SA was so popular in 1950s?
  5. Is SA a single objective or a multiojective optimization algorithm?
  6. How would you terminate a stochastic algorithms such as SA sooner?
  7. What is mutation? Give an example.
  8. What type of surfaces(objective spaces) will SA find hard to optimize?
  9. What is a metaheurestic? State 2 properties of meta-heurestics.
  10. When finding a solution, you can either mutate towards ''Heaven'' (A better spot) or you can choose to mutate away from "Hell" (A worse spot). Why would you choose one over the other? (HINT: One of them has a better diversity of search.)

Practice

  1. In about 10 lines, write down the pseudo-code for SA. Number each line.
  2. In the pseudo-code for SA, you used a neighbourhood function Neighbour(). Code up the Neighbour function.
  3. In the pseudo-code for SA, you used a probability function P(e_new, e_old, t). What would be a valid mathematical expression for this? How would you change this for a maximization problem?