Courses' slots:
Week one 17:00 - 18:30
Language and Logic advanced course:
Theoretical and Empirical Perspectives on Discourse.
Teacher
Abstract: This course will present an overview of current research on discourse both from a theoretical and empirical point of view. I will begin with the theoretical point of view, concentrating on Segmented Discourse Representation Theory (SDRT), which has made most explicit the ties between discourse semantics and more traditional linguistic fields of pragmatics and semantics. I will also compare RST, DLTag and other theories to SDRT. I then will detail the empirical stumbling block for all of these theories, which is the scaling up to real world, domain independent texts. I will review empirical approaches to discourse parsing in an attempt to address this empirical problem. I will discuss corpus construction and learning methods, leaning on the results of the Penn Discourse Tree Bank, Graph Bank, and the SDRT annotation projects DISCOR and ANNODIS.
Logic and Computation workshop:
Structures and Deduction 2009.
Organizers
- Michel Parigot ()
- Lutz Strassburger ()
Course material: structures_deduction.pdf
Abstract:
The topic of this workshop is the application of algebraic,
geometric, and combinatorial methods in proof theory. In
traditional proof theory, research is focussed on syntax, but in
recent years many researchers have proposed approaches to avoid
"syntactic bureaucracy" in the presentation of proofs. Examples
are proof nets, atomic flows, new deductive systems based on
deep inference, and new algebraic semantics for proofs. These
efforts have also led to new methods of proof normalisation and
new results in proof complexity.
Thus the workshop is relevant to a wide range of people. The
list of topics includes among others: algebraic semantics of
proofs, game semantics, proof nets, deep inference, tableaux
systems, category theory, deduction modulo, cut elimination,
complexity theory, computational interpretation of proofs, etc.
The goal of the workshop is twofold: first, to bring together
researchers from various fields who share the interest in
reducing the dependency of logic from low-level syntax, and
second, to provide an opportunity for PhD students and
researchers to present and discuss their work with colleagues
who work in the broad subject areas that are represented at
ESSLLI.
The workshop is intended to be a sequel of the ICALP-workshop
SD05 in Lisbon 2005 .
Language and Logic introductory course:
Logicality and Invariance.
Teacher
- Denis Bonnay ()
Course material: bonnaylogicalityandinvariance.pdf syllabuslogicalityandinvariance.pdf vanbenthembonnaymodallogicinvariance.pdf
Abstract:
When designing a logical system, one has to start by distinguishing between logical expressions, which are the proper objects under study and whose interpretations remain fixed across models, and non-logical expressions, whose interpretations are permitted to vary from one model to the other. In particular, this distinction is crucial for the definition of logical truth as truth in all models. But are there principled ways to draw the line between logical and non-logical expressions? Consider any logical system, what is its natural repertoire of logical operations?
The aim of the course will be to review the semantic answers to these questions and to evaluate their consequences with respect to general questions in logic (such as functional completeness results), in linguistics (such as the semantic nature of `functional' or `grammatical' expressions) and in the philosophy of the formal sciences (why is logic different from mathematics?).
I will present the standard approach in terms of permutation invariance, which is generally credited to Tarski, and then introduce a generalized framework for the analysis of various invariance criteria. In particular, I will show how bisimulation invariance and invariance under potential isomorphism come out as natural notions of invariance for modal logic and first-order logic.
Course materials will be made available on my webpage:
http://lumiere.ens.fr/~dbonnay/
Logic and Computation introductory course:
Games, Action and Social Software.
Teachers
Abstract:
Social software is a term coined by Rohit Parikh for the emerging discipline that investigates the logical, computational and strategic aspects of social mechanisms. The course will give an overview of choice topics in this area, and draw connections with game theory, action logic, epistemic logic, and social choice theory.
Point of departure will be a collection of Discourses on Games, Action and Social Software that is due to appear with Amsterdam University Press, in the Texts in Logic and Games series.
The level of the course will be the level of the discourses in the book, or slightly more advanced.
Each course lecture will be based on one or two of the discourse chapters from the book.
Language and Computation advanced course:
Computational Psycholinguistics.
Teacher
- Roger Levy ()
Over the last two decades, computational linguistics has been revolutionized by increases in computing power, large linguistic datasets, and a paradigm shift toward the view that language processing by computers is best approached through the tools of statistical inference. During roughly the same time frame, there have been similar theoretical developments in cognitive psychology towards a view of major aspects of human cognition as instances of rational statistical inference, exemplified by work such as Anderson (1990) and Tenenbaum & Griffiths (2001). Developments in these two fields have set the stage for renewed interest in computational approaches to human language processing. Correspondingly, this course covers some of the most exciting developments in computational psycholinguistics over the past decade. The course focuses on probabilistic knowledge and memory in language processing, covering models, algorithms, and key empirical results in the literature.
Language and Logic workshop:
Vagueness in Communication.
Organizers
Course material: vic09-alxatib-pelletier.pdf vic09-daniel-lassiter.pdf vic09-ewan-klein.pdf vic09-gottfried-vosgerau.pdf vic09-harald-bastiaanse.pdf vic09-lavi-wolf.pdf vic09-pablo-cobreros.pdf vic09-paul-egre.pdf vic09-stephanie-solt.pdf vic09-zardini-sweeney.pdf
Abstract:
Although vagueness has long since been an important topic in
philosophy, logic and linguistics, some recent advances have made the
functions of vagueness in natural language communication an exciting
and timely research area. This renewed interest has a distinct
cross-disciplinary character and has spawned many new research
questions. While the classical instruments of dealing with vagueness
-- like multi-valued logics, truth value gaps or gluts, or
supervaluations -- have not been significantly extended, new
approaches investigate questions like context-sensitivtiy of vagueness
(Kyburg & Moreau 2000), the sharpening of vague predicates in context
(Barker 2002), and the modeling of precision levels with expressions
like 'roughly' or 'like' (Kennedy 2007). Within the study of
comparatives and degree modification, moreover, researchers are now
exploring cross-linguistic aspects of vagueness (Beck et al 2004). On
a more fundamental level, the question why there is vagueness to begin
with, what role vagueness serves in human communication, has been
addressed. For example, it has been argued that vagueness is an
epiphenomenon of the impossibility of complete shared knowledge about
the extension of many terms (e.g. Williamson 1994), hence a
consequence of the cognitive limitations of humans. It has been shown
why this does not affect the utility of these terms in communication
(Parikh 1994). Game-theoretic methods have been employed that show
that being vague or imprecise can be beneficial for communication even
if the speaker could truthfully use more precise terms (de Jaegher
2003). Furthermore, the important role of vagueness became evident in
a number of empirical domains beyond obvious examples such as the
language of diplomacy -- for example, in geographical terms (e.g.
Bennett 2008) or in the description of measures of economy (Qizilbash
2005). There are also initial experimental investigations into the
ways how speakers interpret vague terms (e.g., Bonini e.a. 1999).
The workshop aims to bring together researchers whose work contributes
to the broad inter-disciplinary line of inquiry outlined here. In
particular, we welcome:
* papers that broaden the empirical base for the study of
vagueness, be it linguistic or otherwise;
* papers offering a synthesis of theories from different disciplines; and
* papers addressing the pragmatics of vagueness.
The workshop aims to provide a forum for researchers (including
advanced PhD students) to present and discuss their work with
colleagues and researchers who work in the broad subject of the
disciplines relevant for vagueness in communication, as represented in
ESSLLI.