@GaryMarcus @ylecun I agree here. The human brain likely has innate priors to learn faster. Infants learn skills in the same order as other infants. This suggests that there exists a meta-learning algorithm that is innate in humans.
@GaryMarcus @ylecun Primates also learn in the same order as human infants up to a certain age. After some threshold age they cannot keep up. This could be due to a resource constraint or a missing meta-learning algorithm.
@GaryMarcus @ylecun There exists an innate meta-learning algorithm for mammals that leads to the sophistication found in primates, dogs, dolphins, whales, elephants, and seals. We need to discover this first before we can know if the human meta-learning algorithm is different.
@GaryMarcus @ylecun Is the mammalian or human meta-learning algorithm different from the birds and octopuses? If it's the same, then this is evidence of the independence on hardware. I don't know if there are studies that have studied this.
@GaryMarcus @ylecun We do know the commonality between octopuses, birds and intelligent mammals. We all know that they have REM states in sleep. Animals with more innate priors (i.e. Ibex) appear not to require additional brain plasticity as evidence by dreaming states.
@GaryMarcus @ylecun The argument of nature vs nurture can be framed in terms of the Computer Science idea of eager vs lazy evaluation. A case of this is known as pre-mature optimization. Brain plasticity allows deferred evaluation which is a case of loose coupling in time.
@GaryMarcus @ylecun Another phrase that is more familiar would be "late binding". In fact, the entire history of computer science consist of innovation of the increase of late binding. medium.com/intuitionmachi…
@GaryMarcus @ylecun In fact, Ockham's Razor which is a popular principle in formulating theories is a subset of the Loose Coupling principle. Loose Coupling is an information dependency principle and the lower the dependency the more flexible the architecture.
@GaryMarcus @ylecun @threadreaderapp unroll please
• • •
Missing some Tweet in this thread? You can try to
force a refresh
