![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
greyaAll You Need is Love and a Good Computational Search Model
...the sequential search process that humans use to select their mates.
“The sequential search problem is tricky because we can’t see forward in time — we don’t know who we might meet in the coming days — and we can’t go backward in time,” he says.
In his ongoing research, Todd uses computer simulations of different search strategies to see how people can solve this problem, and what happens when they do. Building computer models of this and related behaviors, such as how people search for a good parking space, is what puts Todd’s work within the realm of the complex systems group in the IU School of Informatics.
...the sequential search process that humans use to select their mates.
“The sequential search problem is tricky because we can’t see forward in time — we don’t know who we might meet in the coming days — and we can’t go backward in time,” he says.
In his ongoing research, Todd uses computer simulations of different search strategies to see how people can solve this problem, and what happens when they do. Building computer models of this and related behaviors, such as how people search for a good parking space, is what puts Todd’s work within the realm of the complex systems group in the IU School of Informatics.
no subject
Date: 2005-12-13 07:13 am (UTC)В книгах по flow control есть понятие "оптимизация без цели" - рассматриваем, какие величины вообще можно отрегулировать, меняя процесс управления, а потом уже решаем, какой набор показателей считать оптимальным. Это оно?
"Скажите, поручик, а не получали ли вы ошибки ввода-вывода?" (с)
no subject
Date: 2005-12-13 07:17 am (UTC)(пауза. Голос из аудитории)
"Глубокая мысль. Ээ... то есть, если Вы ведете себя так, как вели бы себя, если бы знали прикуп, то Ваш образ жизни не отличается от того, который бы Вы вели, если бы жили в городе Сочи"
no subject
Date: 2005-12-13 03:46 pm (UTC)There's not enough information to be sure, but I think it's not quite this - mostly because he doesn't seem to be dealing with optimization of search parameters so much as with trying to satisfy some culture-dependent "presets" of these parameters by going through a sequence of people. The article talks about change of partners, but not change of assessment. And then the mention of marriage demographics... I think he's modeling static "average for group x" search strategies.
I like the idea though. Complex enough psychology on top of complex enough math can give interesting results to compare with the real world :)