Epstein Secrets

preschool Minecraft IQ tests, helps in terms of conferring ability to pass the robot preschool IQ tests... One thing that would interest me -- and Joscha -- would be working toward embodied agents that could pass these IQ tests "via learning language in an embodied way" .... This might not be the sho

ar old can do. >» the challenge is learning a language. different that moving blocks in »>a >» video game. »> >» On Sun, Sep 6, 2015 at 2:38 PM, Joscha Bach >» wrote: >> >> » » This challenge idea is excellent; I really love it! » » » >» first draft. of the Chomsky Challenge. . Produce a non- living

mulate or understand learning language like a baby, I suspect it's necessary to go robotic (though, certainly, not necessarily HUMANOID-robotics). Joscha may disagree on this point, I'm unsure.... The point is you need a rich stream of perceptual data, whose interrelationships can ground the interrel

» complex structures. The winning approach would then not be able to » work on another language for which there was no large annotated » corpus (no Penn Treebank analogue, etc.). » It seems that challenges are easier to formulate for engineering » breakthroughs than science breakthroughs... » Here is one id

preschool Minecraft IQ tests, helps in terms of conferring ability to pass the robot preschool IQ tests... One thing that would interest me -- and Joscha -- would be working toward embodied agents that could pass these IQ tests "via learning language in an embodied way" .... This might not be the sho

ar old can do. >» the challenge is learning a language. different that moving blocks in »>a >» video game. »> >» On Sun, Sep 6, 2015 at 2:38 PM, Joscha Bach ca >» wrote: >> >> » » This challenge idea is excellent; I really love it! » » » >» first draft. of the Chomsky Challenge. . Produce a non- livin

mulate or understand learning language like a baby, I suspect it's necessary to go robotic (though, certainly, not necessarily HUMANOID-robotics). Joscha may disagree on this point, I'm unsure.... The point is you need a rich stream of perceptual data, whose interrelationships can ground the interrel

» complex structures. The winning approach would then not be able to » work on another language for which there was no large annotated » corpus (no Penn Treebank analogue, etc.). » It seems that challenges are easier to formulate for engineering » breakthroughs than science breakthroughs... » Here is one id

From: on behalf of Ben Goertzel Sent: Monday, September 7, 2015 12:17 AM To: Jeffrey E. Cc: Joscha Bach Subject: Re: Hi, I agree w/ Joscha's caution about discrimination tasks: They can be often be solved rather well, but in devious ways, by statistical supervised learning algorithms.

From: on behalf of Ben Goertzel Sent: Monday, September 7, 2015 12:17 AM To: Jeffrey E. Cc: Joscha Bach Subject: Re: Hi, I agree w/ Joscha's caution about discrimination tasks: They can be often be solved rather well, but in devious ways, by statis

quite complex structures. The winning approach would then not be able to work on another language for which there was no large annotated corpus (no Penn Treebank analogue, etc.). It seems that challenges are easier to formulate for engineering breakthroughs than science breakthroughs... Here is one idea, off

ules.. babys learn la=guage. many of them . figure out how the baby does it. On Sun, Sep 6, =015 at 8:16 PM, Ben Goertzel > wrote: Hi, I agree w/ Joscha's caution about discrimination tasks: They can be often be solved rather well, but in devious ways, by statistical supervised learning algorithms.

ngs a 2 year old can d=. > the challenge is learning a language. different that=moving blocks in a > video game. > On Sun, Sep 6, 2015 at 2:38 PM, Joscha Bach > wrote: >> » This challenge idea is excellent; I really love it! >> >» first draft. of the Chomsky Challenge. . Produce a non- living >» sys

e=br> complex structures. The winning approach would then not be able=to work on another language for which there was no large annotated corpus (no Penn Treebank analogue, etc.). It seems that challenges are easier to formulate for engineering breakthroughs than science breakthroughs... Here is one idea, off

From: on behalf of Ben Goertzel < Sent: Monday, September 7, 2015 12:17 AM To: Jeffrey E. Cc: Joscha Bach Subject: Re: Hi, I agree w/ Joscha's caution about discrimination tasks: They can be often be solved rather well, but in devious ways, by statistical supervised learning algorithms.

quite complex structures. The winning approach would then not be able to work on another language for which there was no large annotated corpus (no Penn Treebank analogue, etc.). It seems that challenges are easier to formulate for engineering breakthroughs than science breakthroughs... Here is one idea, off