Saturday, March 24, 2007

The Mechanical Turk and Searle's Chinese Room

The Times has an article about Jeff Bezos' Mechanical Turk project, which lets machines outsource certain tasks to humans. (The orginal mechanical Turk was an 18th century hoax in which a hidden human operated a chess-playing automaton.) As Bezos describes,

“Normally, a human makes a request of a computer, and the computer does the computation of the task,” he said. “But artificial artificial intelligences like Mechanical Turk invert all that. The computer has a task that is easy for a human but extraordinarily hard for the computer. So instead of calling a computer service to perform the function, it calls a human.”

...The company opened Mechanical Turk as a public site in November 2005. Today, there are more than 100,000 “Turk Workers” in more than 100 countries who earn micropayments in exchange for completing a wide range of quick tasks called HITs, for human intelligence tasks, for various companies.

The Times writer Jason Pontin (who is also editor and publisher of MIT's Technology Review), gives Turk working a try, and finds it disorienting:

What is it like to be an individual component of these digital, collective minds?

To find out, I experimented. After registering at www.mturk.com, I was confronted with a table of HITs that I could perform, together with the price that I would be paid. I first accepted a job from ContentSpooling.net that asked me to write three titles for an article about annuities and their use in retirement planning. Then I viewed a series of images apparently captured from a vehicle moving through the gray suburbs of North London, and, at the request of Geospatial Vision, a division of the British technology company Oxford Metrics Group, identified objects like road signs and markings.

For all this, my Amazon account was credited the lordly sum of 12 cents. The entire experience lasted no more than 15 minutes, and from my point of view, as an occluded part of the hive-mind, it made no sense at all.

This is reminiscent of philospher John Searle's thought experiment called the Chinese Room, in which he posits a large team of humans implementing an algorithm that translates Chinese to English. Since each human performs only a small task (e.g., sorting acording to a rule set), none have any understanding of the overall process. Searle asks where, exactly, does the understanding of Chinese and English reside in this device? Searle considered his thought experiment as evidence against strong AI, whereas I just consider Searle to be confused. It's obvious that a Turk worker might be a small cog in some larger process that "understands" the world and processes information in a useful way. This depends not at all on what the little cog understands or does not understand.

5 comments:

Unknown said...

See The Mechanical Turk as well.

Steve Hsu said...

Mark,

Thanks for the link -- your post has some interesting detail as to how the process works!

Steve

Unknown said...

The difference is - if it's divided up enough that you can't point to a person or persons who are doing translation, then Searle's machine probably wouldn't work any better than machine translation. If you break the whole process into very small steps that can be written down for people to do, you're moving closer and closer to machine translation implemented with people. Since machine translation doesn't work as well (yet?) as machine + person, it's pretty easy to point to who's doing what part of the translation work. Isn't it?

Steve Hsu said...

J,

Any Chinese Room implementation today would clearly be deriving much benefit from the humans inside. But that doesn't mean that the task won't eventually be broken down into smaller chunks (better machine translation). The question is whether you think that is impossible, or whether you think Searle's thought experiment *shows* that is impossible. I do not think it does.

The point about the MTurk is that there clearly are applications of it with the following properties:

1) it processes information in a useful way, so that it must "understand" (encode) something abot the world

2) the humans inside don't understand the big picture of what is happening

These two together are a kind of refutation of Searle's argument, albeit not for the specific task of language translation.

Margaret Rouse said...

I've been watching Mechanical Turk for awhile and thought it was a big deal about nothing...until I learned about Powerset. Now I'm convinced the two are connected and Mechanical Turk is the secret ingredient that's going to allow Search Wikia to challenge Google's throne.
Could I be right?

Blog Archive

Labels