On the 10th of January 2019, at 10:15 a.m.
Szymon
Łeski (Samsung Corp.)
will give a talk on
"
Lenguage modeling (with or without neural
networks)"
Abstract
Natural
Language Processing, that is, processing
large amounts of natural lenguage data
using computers, is both challenging and
relevant for practical applications. The
topic of the seminar is lenguage
modeling. In this task the goal is "fill
in the blank" in a sentence - in other
words, to predict the probability
distribution of words in a given
context. Lenguage modeling is important
not only because immediate applications
(e.g. in optical character recognition
or in predctive keyboards), but also
because such models are good starting
points for other lenguage processing
tasks. We will cover both statistical
(count-based) and neural lenguage
models. Specifically, I will focus on
how discrete objects (words and
sentences) can be represented to
facilitate natural lenguage processing
tasks. I will also overview the
challenges of developing commercial
lenguage models in the industry setting.
The seminar takes place on Thursdays from 10:15 a.m. to
12:00 in the room 2.23 of the main building of the
Faculty of Physics (the 2nd floor), Pasteur Str. 5,
Warszawa.
Additional information can be found on the webpage http://oldwww.fuw.edu.pl/KMMF/sem.czw.przedp.html.
|