Polyworld: Using Evolution to Design Artificial Intelligence

Google Tech Talks
November, 8 2007

ABSTRACT

This presentation is about a potential shortcut to artificial intelligence by trading mind-design for world-design using artificial evolution. Evolutionary algorithms are a pump for turning CPU cycles into brain designs. With exponentially increasing CPU cycles while our understanding of intelligence is almost a flat-line, the evolutionary route to AI is a centerpiece of most Kurzweilian singularity scenarios. This talk introduces the Polyworld artificial life simulator as well as results from our ongoing attempt to evolve artificial intelligence and further the Singularity.

Polyworld is the brain child of Apple Computer Distinguished Scientist Larry Yaeger, who remains the primary developer of Polyworld:

http://www.beanblossom.in.us/larryy/P...

Speaker: Virgil Griffith
Virgil Griffith is a first year graduate student in Computation and Neural Systems at the California Institute of Technology. On weekdays he studies evolution, computational neuroscience, and artificial life. He did computer security work until his first year of university when his work got him sued for sedition and espionage. He then decided that security was probably not safest field to be in and he turned his life to science.

• Category

  People & Blogs

[RegularSmith]

Can everyone stop being babies. The kid is obviously very excited about his work and has a fixed time schedule to adhere to. People are such bitc*** these days.

[gattocake]

"Our speaker is Virgil Griffith. He's talking about the theory of relativity."

[lennyhome]

At 5:00 you can see proof that macs never crash.

[DjDedan]

did he say that? perhaps he meant once a nervous system evolved? But i guess how would he prove that... still interesting stuff...

[DjDedan]

a little confused about the experiment where he turns off evolution - aren't the genes that are randomly used still a result of selection? That is, you can only randomly select from creatures still alive - thus selected. There may be a scale difference in complexity but it's still evolution. Perhaps simply cloning with no cross-over or mutations is evolutionless?

[Seanathon Balkmenistan]

Relativity is a lot easier than this if you're good at math.

[Randy Olson]

Sounds like something worth pursuing as part of a graduate program! Hop to!

[WSWarthog]

Well then for your sake I hope I'm wrong about its being a dead end.

I have my own ideas of a quasi-evolutionary approach where instead of evolving whole organisms you attempt just to evolve a cognitive structure capable of performing well on g-loaded tasks, using a specific grammar defining how these cognitive structures grow (not unlike the grammar of l-systems).

But it's just a notion, no idea of how viable it would be.

[Randy Olson]

It's only the focus of my PhD research. ;-)

[WSWarthog]

It sounds like you're somewhat more familiar with the subject than me, so I'll have to concede.