Re: Cangelosi/Harnad Symbols

From: Jelasity Mark (jelasity@amadea.inf.u-szeged.hu)
Date: Mon Dec 27 1999 - 19:30:33 GMT


I don't have too much to add to S.H.'s comments on my points regarding
the target article.

However, I have read the other article on grounding transfer. I liked
this one much better. The model presented here is much cleaner,
simpler, and yet more sophisticated: the perceptual input looks like
perceptual input, and it has no structural correlation with signs.
And, no question, within this model there is grounding transfer.
I missed a more rigorous analysis of this phenomenon, but the authors
did too, as they state in the Discussion.

Though its relation with the development of language (and in particular
with Harnad's theory) is still an open question to me, this paper poses
a nontrivial and interesting research topic: what are the conditions
that allow grounding transfer in this family of models? My first
impression is that probably the most important condition is the "right"
grounding of basic categories in the toil phase. This means that the
names should trigger the same inner representation (here: hidden unit
activation) than perceptual input. The conditions of this condition are
still non-trivial though, since the space of hidden unit
representations can have a real difficult structure especially if there
are many hidden units (in this paper there are 55 input units, but only
5 hidden units).

If the CONDITIONS of this grounding transfer turn out to be compatible
with some biologically and evolutionary plausible story of the
development of mental representations, the model may have a really
large impact, unlike I suggested based only on the target article.

In the following I reply to two comments that are not very closely
related to the target article.

sh> jm> I only said here, that once you have some
sh> jm> categories grounded, then you can use them as input to learning higher
sh> jm> level categories. This is also easier than pure honest toil, it is a
sh> jm> sort of self-theft, though the "name" of the target category is of
sh> jm> course missing. The type of learning is irrelevant.
sh>
sh> I'm not sure what you mean. You can't learn a new category by talking to
sh> yourself (before you have language!). If you are thinking of reasoning,
sh> it is a little premature in this model. Same for Vygotskyian inner
sh> speech. We need to get to outer speech before we can get to inner
sh> speech....

The process I'm referring to does not even have to be conscious. The
categories learnt via toil "are there". I mean (again a further theory
of mine) that they have to have some kind of localist representation
(not necessarily physically localist), just like the neurons in the
retina firing in the presence of certain shapes, and these locally
represented categories can serve as input to further learning (via
toil). Higher level (but not linguistic, only "reverse engineered")
categories do not receive the most basic input directly; they may be
based on lower level categories.

sh> jm> I'd go to the zoo, and I'd took a good look at the animal which has
sh> jm> "ban-ma" written on its cage. Language works like God, who provides
sh> jm> names and helps us learn to ground them from others who already know
sh> jm> their meaning.
sh>
sh> Ah, this is similar to your point about the possibility of learning a
sh> category from positive instances alone (and suffers from the same sort
sh> of problem, namely, that in nontrivial cases it is impossible).
sh>
sh> Yes, If all members of a category, besides their critical features
sh> (which are hard to find, and normally need to be learnt by honest toil)
sh> also wore their names on their sleeves, then categorization would be a
sh> lot easier (indeed, you would not have to worry about figuring out
sh> features at all). Indeed this is precisely what cheating is.
sh>
sh> No, in a nontrivial category learning task, my telling you that THIS is
sh> a ban-ma would do you next to no good in deciding whether or not the
sh> next candidate was a ban-ma too (just as eating one mushroom, and
sh> either not getting sick, does not thereby make me capable of
sh> distinguishing the edible from the inedible mushrooms). It's not just a
sh> matter of being given a positive instance and contemplating it till its
sh> critical features leap out at you. The critical features are detected
sh> by trial and error, from sampling many positive and negative
sh> instances (with the help of an internal implicit learning device --
sh> possibly a neural net -- that it good at doing just that).

This subject leads very far, just a short comment.

The role of the examples depends on the learning algorithm. There is a
mathematical finding, that is relevant here: every learning algorithm
has a BIAS (including backprop on NNs). Otherwise any generalization to
unseen examples is impossible. The bias means that the learner prefers
some categories A PRIORI. If this bias is correct, then the learning
algorithm is successful, otherwise it isn't.

If the bias is such that specific categories (i.e. representing a small
subset) are preferred, then usually only positive examples are
interesting.
If the bias is towards general concepts, then only negative examples
are interesting.

I don't understand exactly what you mean on "trivial" tasks, but the
role of positive examples completely depends on the nature of
categories to be learned i.e. the correct bias for the (maybe abstract)
domain. I don't see why domains where only positive examples are needed
should necessarily be more trivial then other domains, whatever that
means.



This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:07 GMT