Comments on: Growth Mode Jump Example

By: mjgeddes

mjgeddes — Tue, 16 Jun 2009 03:58:09 +0000

ShardPhoenix,

I’ve solved the puzzle of reflection in AI that long stumped … er certain other people… for so long. I will post a short abstract of my solution in next month’s Open Thread, publically proving once and for all that *my* Reflection trumps *his* IQ. Robin, that next post in the Open Thread will be my last on this blog – promise

By: Tim Tyler

Tim Tyler — Sun, 14 Jun 2009 14:59:45 +0000

The traditional timeline looks something like this:

* Hunting & Gathering;
* Herding 15,000 BC;
* Agricultre 10,000 BC;
* Urban 8000 BC – cities form;

Agriculture and group living seem fairly strongly associated with each other. Living in groups allowed humans to specialise – and move away from the shaman-hunter-gatherer-chief model into one where there were many different roles in society.

By: ShardPhoenix

ShardPhoenix — Sun, 14 Jun 2009 12:23:47 +0000

The guy was agreeing with you (I think) and constrasting your post to what he thinks Eliezer believes. And don’t be so jealous of Eliezier stealing your thunder :/.

By: ShardPhoenix

ShardPhoenix — Sun, 14 Jun 2009 12:22:36 +0000

Elizezer is all about reflection, and I don’t think he’s ever been obsessed with “IQ” in the shallow way you’re accusing him of here.

By: Mario

Mario — Sun, 14 Jun 2009 06:13:31 +0000

I think that technological and cultural innovation receives its main support from the luxury of leisure. Or, I suppose more accurately, the time to focus on things other than the immediate. If this were true, the first innovations would have begun simply due to economies of scale, which is what I think this shows. I’m not sure what precipitated the agricultural revolution, but I don’t think it could have happened had people not been secure enough in their situation to wait in one place for months for their main food supply to mature. You would expect similar boosts from the invention of slavery and the industrial revolution, as the returns to small inputs in labor became high enough to support more and more people working outside of the survival sector (although I’m not sure the latter is possible in a society until the former is removed).

By: Michael F. Martin

Michael F. Martin — Sun, 14 Jun 2009 03:14:57 +0000

Nice affinities with recent work of Axelrod. Some version of Kuramoto model could be used here also.

By: Robin Hanson

Robin Hanson — Sun, 14 Jun 2009 02:49:24 +0000

evil, this is my person blog, and signaling is one of my interest areas. Why should I not focus my post topics on my interest areas?

By: Robin Hanson

Robin Hanson — Sun, 14 Jun 2009 02:46:50 +0000

mjgeddes, perhaps it escaped your attention that I not Eliezer was the author of the post, which had nothing to do with IQ? Eliezer doesn’t even post here at OB anymore. Perhaps you could take your obsession with him somewhere else?

By: mjgeddes

mjgeddes — Sun, 14 Jun 2009 02:26:59 +0000

What this tells us once again is that Eliezer Yudkowsky is wrong about everything. IQ is not the most powerful force in the universe. It’s the interaction/communication/signalling between agents that resulted in the explosion of innovation, not any increase in raw IQ.

And what cognitive ability does interaction/communication/signalling involve? Clearly, the ability to form analogies (mappings) between between concepts – the core aspect of communication, i.e. case-based reasoning.

The most powerful force in the universe is reflection (the ability to translate between and communicate different knowledge representations), not intelligence. It’s actually this reflective/communicative/signalling ability that’s at the heart of the growth mode jumps.

By: TGGP

TGGP — Sat, 13 Jun 2009 23:43:54 +0000

John Hawks discusses the paper here.