1 / 5

Alternate Computational Forms

Alternate Computational Forms. We have found cases where... Solution space is very large, O(2 n ), O(n!) No algorithmic method to efficiently find the solution can be found The problem is too important to ignore What to do? Use techniques used to approximate solutions.

odin
Download Presentation

Alternate Computational Forms

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. Alternate Computational Forms • We have found cases where... • Solution space is very large, O(2n), O(n!) • No algorithmic method to efficiently find the solution can be found • The problem is too important to ignore • What to do? Use techniques used to approximate solutions. • Heuristic solutions ( “Rule of thumb”) • Semi-random search through solution space • Non-optimal solutions found • Basic Dilemma: Balancing randomness required to explore solution space, with stability required to retain effective candidates. 11

  2. Genetic Algorithms – Overview • Represents potential solutions as long strings, termed “genomes.” • More “fit” solutions are evolved as various genomes are tried and evaluated. • Higher rated (fit) genomes are allowed to survive and “reproduce” with other fit genomes. 01110000110100011001101 01110000110100011001101 01110110110100000100001 10110111001010001000011

  3. Genetic Algorithms – Basic Sequence • Create initial population of random solutions • Evaluate fitness of each member of the population. • Use most fit members of the current generation to create new potential solutions: • Selection: Select two or three “parents.” • Crossover: Combine genetic material by copying from one parent, crossing over to the other parent periodically. • Mutation: Randomly modify a small subset of the new solution. • Acceptance: Incorporate the new solutions into the population (usually removing less fit members). • Repeat until fitness achieves some threshold or no longer increases.

  4. Genetic Algorithms - Links • “Hello World!” • http://www.generation5.org/content/2003/gahelloworld.asp • Examples with Applets • http://cs.felk.cvut.cz/~xobitko/ga/ • More examples • http://www.mathworks.com/access/helpdesk/help/toolbox/gads/index.html?/access/helpdesk/help/toolbox/gads/f6691.html&http://www.google.com/search?hl=en&q=%22genetic+algorithm%22+example

  5. Simulated Annealing • Based on annealing of metal: • High temperature transforms metallic crystals and allows them to bond randomly • Slowly lowering the temperature can allow the crystals to “freeze” in a preferred way. • Potential solutions that are “neighbors” of the current one are attempted randomly. • The “distance” between solutions is initially large, and grows smaller as the temperature “cools.”

More Related