Basically, the algorithm will let computers learn from their virtual experiences by measuring the distance between a desired outcome and an actual outcome. Using this measurement, Professor Mansour hopes that the computer will be able to predict the future and minimize future regret.
“If the servers and routing systems of the Internet could see and evaluate all the relevant variables in advance, they could more efficiently prioritize server resource requests, load documents and route visitors to an Internet site, for instance,” Professor Mansour says.
And this is where Google comes in–not only is Mansour’s vision efficient (Google is all about efficiency), but Google hopes to use Mansour’s research to improve AdWords and Adsense.
“Compared to human beings, help systems can much more quickly process all the available information to estimate the future as events unfold–whether it’s a bidding war on an online auction site, a sudden spike of traffic to a media website, or demand for an online product,” Mansour says.
Hopefully, the algorithm will update itself as it runs and “adapt to the situation at hand.” Mansour suggests that, after a task is finished the results will be “almost as if you knew all the variables in advance.”
So yeah, it’s not really regret, per se. Rather, computers will be able to look back on undesired outcomes, determine the difference between that outcome and the desired outcome, and predict what will happen in the future (based on this experience) and take steps to make the next outcome closer to the desired outcome.
For lack of a better word, Mansour calls it “regret.” I do hope that computers are better at the “learning” part of the process than are, say, college students.
[Tel Aviv University via Engadget]
Follow Sarah on Twitter (@ geeklil ) or on Facebook .
Like this? You might also enjoy…
- IBM’s Watson Not as Smart as You Think
- Medical Robot Plays ‘Operation,’ Still Needs Human Help
- Robots Can Juggle Better Than You
Get your GeekTech on: Twitter – Facebook – RSS | Tip us off