In a recent debate between Gary Marcus and
Yoshua Bengio about the future of Artificial Intelligence, the question came up
of how much information the genome can encode. This relates to the idea of how
much innate or prior “knowledge” human beings are really born with, versus what
we learn through experience. This is a hot topic in AI these days as people
debate how much prior knowledge needs to be pre-wired into AI systems, in order
to get them to achieve something more akin to natural intelligence.
Bengio (like Yann leCun) argues for putting
as little prior knowledge into the system as we can get away with – mainly in
the form of meta-learning rules, rather than specific details about specific
things in the environment – such that the system that emerges through deep
learning from the data supplied to it will be maximally capable of generalisation.
(In his view, more detailed priors give a more specialised, but a more limited and
possibly more biased machine). Marcus argues for more…
we’ve been thinking about the genetics of intelligence from completely the
wrong angle? Intelligence (as
indexed by IQ or the general intelligence factor “g”) is clearly highly heritable in humans – people who are more genetically
similar are also more similar in this factor. (Genetic variance has been estimated as explaining ~75% of variance in g,
depending on age and other factors).
There must therefore be genetic variants in the population that affect
intelligence – so far, so good.
But the search for such variants has, at its heart, an implicit
assumption: that these variants affect intelligence in a fairly specific way – that
they will occur in genes “for intelligence”. An implication of that phrase is that mutations in those genes were positively
selected for at some stage in humanity’s descent from our common ancestor with
apes, on the basis of conferring
increased intelligence. This
seems a fairly reasonable leap to make – such genes must exist and, if variation
Mental illness is surprisingly common.About 10% of the population is affected by it at any one time and up to 25% suffer some kind of mental illness over their lifetime.This has led some people (many people in fact) to surmise that it must exist for a reason – in particular that it must be associated with some kind of evolutionary advantage.Indeed, this is a popular and persistent idea both in scientific circles and in the general public.(See the recent article “Depression’s Upside” from the New York Times Magazine, for example).Such theories come in two main varieties – the first, that mental illness confers some specific advantage to those afflicted; and second, that the mutations which cause mental illness in one person’s genetic background may confer an advantage when they are in a different genetic background (balancing selection).Both of these suffer from some misconceptions about how evolution by natural selection works.The intuitive appeal of the “survival of the fittest” meta…