Perspectives on Language and Intelligence

sounderslipInternet and Web Development

Oct 22, 2013 (3 years and 7 months ago)

55 views

Perspectives on Language and
Intelligence

Giuseppe Attardi

Dipartimento di Informatica

Università di Pisa

Università di Pisa

Language and Intelligence

“Understanding cannot be measured by
external behavior; it is an internal metric
of how the brain remembers things and
uses its memories to make predictions”.


“The difference between the intelligence of
humans and other mammals is that we
have language”.




Jeff Hawkings, “On Intelligence”, 2004

Mountcastle observation

All the different regions of the neocortex
look pretty much exactly the same.

To understand spoken language, scientists
build algorithms based on rules of
grammar, syntax, and semantics.

But if Mountcastle is correct, the algorithm
of the cortex must be expressed
independently of any particular function or
sense.

The brain uses the same process to see as
to hear. The cortex does something
universal that can be applied to any type
of sensory or motor system.

Hawkins’ Memory
-
Prediction
framework


The brain uses vast amounts of
memory to create a model of the
world. Everything you know and
have learned is stored in this model.
The brain uses this memory
-
based
model to make continuous
predictions of future events. It is the
ability to make predictions about the
future that is the crux of intelligence.

In Summary


Prediction is the primary function of the
neocortex.


All behavior is a byproduct of
prediction.


The neocortex is uniform.


The neocortex stores sequences of
patterns.


The patterns stored in the neocortex can
be recalled in an auto
-
associative way.


The patterns stored in the neocortex must
be stored using an invariant
representation.


Perceptual processing is hierarchical.

More …

“Spoken and written words are just patterns
in the world…

The syntax and semantics of language are
not different from the hierarchical
structure of everyday objects.

We associate spoken words with our
memory of their physical and semantic
counterparts.

Through language one human can invoke
memories and create next justapositions
of mental objects in another human.”

Themes

Vision


David Lowe: Local Invariant Feature


http://www.cs.ubc.ca/spider/lowe/resear
ch.html


Music


Music Genoma Project


http://www.pandora.com/mgp.shtml

Movement


Alain Berthoz. Sense du Movement


Perception and cognition are inherently
predictive, functioning to allow us to anticipate
the consequences of current or potential actions.
The brain acts like a simulator that is constantly
inventing models to project onto the changing
world, models that are corrected by steady,
minute feedback from the world. We move in the
direction we are looking, anticipate the trajectory
of a falling ball, recover when we stumble, and
continually update our own physical position, all
thanks to this sense of movement.

Language


Parser technologies

Ontologies


Omega Description Logic (1979)


Assume that knowledge is represented in
conceptual taxonomies rather than flat
predicates as in FOL


Single
subsumption

relation
is
:


(
a

Man)
is

(
a

Mortal)


Socrates
is

(
a

Man)


Taxonomic reasoning


Evidence from psychometric studies


Danny Hillis planned to use Omega on
Connection Machine


Description Logics

Ontologies


Problem: how to build ontologies?


Hand craft


Acquire from learning


Semantic Web


RDF: derived from description logics


OWL (= DAML+OIL): Web Ontology
Language

Personal stance


Skeptical of Semantic Web approach,
for several reasons:


People


Goals


impractical


Favor approaches based on learning
through large corpora and
continuous adaptation


Alternative View

Cognitive Web based on the Memory
-
Prediction Framework

1.
invariant forms: documents

2.
broad connectivity: the Net

3.
nested feedback loops:aggregation

Knowledge Extraction


Text Analytics, Text Mining


Relation Extraction from parse trees


Semantic Role Labeling


Intent, Opinion Mining


Applications

Opinion Mining


Bibliography at:


http://patty.isti.cnr.it/~esuli/research/sen
timent/

Cognition


Susan Blackmore: Meme Machine


Meme: units of cultural transmission


Role of imitation in humans


Explores the meme
-
gene parallels and derives an
interesting framework for explaining the unusual
size of the human brain and the origins of
consciousness, language, altruism, religion, and
orkut.



http://www.amazon.com/exec/obidos/tg/detail/
-
/019286212X/qid=1091859256/sr=8
-
1/ref=pd_ka_1/103
-
2477453
-
3275828?v=glance&s=books&n=507846

References


A. Berthoz. The brain's sense of
movement. Harvard University Press
.
2000.


J. Hawkins, S. Blakeslee. On
Intelligence. Times Books, 2004.


S. Blackmore. Meme Machine.

Themes for discussion


Lenci: linguaggio e cognizione


Esuli Sebastiani: opinion mining?


Improvvisazione: Ciancia


Neurofisiologia: Felicioli


Semantic Web: Razvan Popescu


Movement: Bonchi


Vision: Scordino


Memes: Passaro


Language:

Mailing List


pangon@gmail.com

Felicioli Claudio


fibonchi@di.unipi.it

Filippo Bonchi


ciancia@di.unipi.it

Vincenzo Ciancia


veraldi@di.unipi.it

Luca Veraldi


carmigna@di.unipi.it

Nicola Carmignani


fiorin@di.unipi.it

Valentino Fiorin


he@di.unipi.it

Yan He


corfini@di.unipi.it

Sara Corfini


popescu@di.unipi.it

Razvan Popescu


scordino@di.unipi.it

Claudio Scordino


nids@di.unipi.it

Francesco Nidito


passaro@di.unipi.it

Alessandro Passaro


callieri@isti.cnr.it

Callieri Marco


baronti@di.unipi.it

Flavio Baronti


ruggieri@di.unipi.it

Salvatore Ruggieri


simi@di.unipi.it

Maria Simi


attardi@di.unipi.it

Giuseppe Attardi