So far in the optimization class, we've solved problems using simulated annealing, genetic algorithms and evolutionary algorithms. This week we begin studying Tabu search. This one is characterized by a list of tabu partial solutions. Yes, tabu, in that they are frowned upon. It uses a somewhat thoughtful process to decide whether to accept a generated solution. I had to laugh at the documentation since it reminded me of the jokes about aging. Here are some of the things to consider when implementing this one:
- Recency (short term memory) "how recently was I here?"
- Frequency (long term memory) "how often have I been here?"
- Quality (aspiration) "how good is being here?"
- Influence (aspiration) "how far away am I from where I have just been?"
And, the aspiration aspects can override the memory aspects. You can decide that it is so good being here, that you don't care if you've been here a lot and it really isn't a good place to be, you're going anyway. I can't wait to write this program!! But, I have to work on linear programming first.
Huge flakes of snow! I can't believe it. So far it isn't sticking, but if it keeps up like this, it might. I'm surprised AU hasn't emailed me to say the campus is closed. Maybe they don't know.
No comments:
Post a Comment