Sunday, June 12, 2016

The Flesh Engine Revs Up


distributed artificial intelligence, aka the swarm

Watch out Skynet, the Flesh Engine is starting to flex its wet muscles. Or should I say - watch out future, because we're about to know the shit out of you.

A swarm AI predicted the Kentucky Derby. The overall field of predictive analysis is taking strides. First there was the guy who predicted the Obama election, over at Nate Silver's 538.

His was all-algorithm though. Good algos, but not this good. As expected, it turns out that people are still better. James Surowiecki, in his book The Wisdowm of Crowds (2004), asserts that large groups of regular people are better at making decisions, and at predictions, than anything. They're better than a small group of experts, and they're better than anything an algorithm can do. As individuals, we're fallible, but as a group, we cross-cancel each other. In Surowiecki's opening example, taken from statistician Francis Galton, a large group of people were asked to guess the weight of an ox. Nobody got it right, but the average of every person's guess was the correct weight.

Surowiecki points out that the stock market predicted the culprit behind the Challenger disaster way before anyone else, because the day after the accident, the stocks in a company that made rubber O-rings went way down. Not until famed scientist Richard Feynman dropped one of those O-rings in a glass of water during a high profile testimony did the groupthinking stock market's "prediction" come true.

China has been plugging in the flesh engine for some time now.

What is the value of living, breathing humans? Stinking, thinking humans? Hopefully we will find out before it's too late, before they're all gone.

Post Script
Yahoo Finance, April 2016

Ahh Headlines



 Mar 24, 2016 report:

(Phys.org)—A female cichlid hybrid fish has been observed to have grown male reproductive organs, impregnate itself and then to have offspring, a team of researchers in the U.K. are reporting in a paper published in the ...

And in other news:

BBC News, March 2016

BBC News, March 2016

Yup.

Finally A User Manual for Being Human


How to be Human
There is a new approach to stalling the takeover of human civilization by Artificial Intelligence. That’s right, we’re going to teach it right from wrong – using crosscultural metamanual moralgorithms.

Taken from phys.org:
“Researchers Mark Riedl and Brent Harrison from the School of Interactive Computing at the Georgia Institute of Technology believe the answer lies in "Quixote"—to be unveiled at the AAAI-16 Conference in Phoenix, Ariz. (Feb. 12 - 17, 2016). Quixote teaches "value alignment" to robots by training them to read stories, learn acceptable sequences of events and understand successful ways to behave in human societies.”

And this is where sh** starts to get real, folks.
“"The collected stories of different cultures teach children how to behave in socially acceptable ways with examples of proper and improper behavior in fables, novels and other literature," says Riedl, associate professor and director of the Entertainment Intelligence Lab.”

feb 2016, phys.org

I’m Not Fkkking Flat You’re Fkkking Flat


image source aplanetruth.info

Because here at network address, we just love all things conspiracy.

Something about gravity waves got the Flat-Earthers mad pissed.

Anyway, I have an idea for a board game where you get ahead by disproving flat earth theories, but instead of a flat game board, it’s a globe, and the first person to get to the center of the earth wins. In the meantime, test the prototype – watch an extensive doc on flat earth conspiracy and try to articulate why the theories don’t add up. Some are easier to disprove than others, but overall, they cover a wide range of disciplines, so there’s a chance for everyone.

gizmodo, feb 2016

Quantum Surrealism


Mutual Metaphors, MARS1, Fecal Face show 2010

Can't wait to see this one come up more often

phys.org, feb 2016

Exciplexes


Exciplexes perhaps

And other words that make me excited, that’s all.

These exciplexes are tunable just by changing the distance between the molecules, even by a few nanometers.

phys.org, feb 2016

Majorana Fermions Strike Again





The Majorana fermion. Love it.

How could you not? The thing is its own anti-particle. So technically, it doesn’t even exist, right?


phys.org, feb 2016