|
Automated
Reasoning
Systems Division
|
| |
SRA Thursday Seminars Series:
2002 (first half)
You are here:
SRA >
Local Seminars >
2002 (first half)
The seminars usually take place on Thursdays at 11:00am in the
SRA Seminars room (Sala Riunioni, room 127, Edificio Est,
seminterrato).
Jump ahead to one of the following dates for details:
Important:
[!] before the link means that the day of the seminar is not
Thursday.
[!] after the link means that one among the location (room) or the
time is not as expected.
Combinations are possible! Please check details with care.
[DT] means "Double Talk", click here
for details.
[DIT] marks the seminars organized in collaboration with the
Department of Information and Communication Technology of the
University of Trento, click here for details.
January 3,
2002
No meeting.
January
10, 2002; 11:00 hrs, [!] room 392 (Sala Conferenze IRST)
Corrado Priami (DIT, Universita` di Trento) will talk about
Global Applications: Design Environments and Case Studies.
Abstract: The talk introduces the main challanges in the
design and development of applications composed
of many autonomous entities that run on a
geographically dispersed network with incomplete
information on the executing environment and with
potentially high rates of faults and disconnection
time. The techniques adopted to cope with this
complexity will be discussed and a way to hide the
formal details to users will also be taken into
account. Some examples of these applications will
be illustrated both in the field of computer science
and biomolecular processes.
SPEAKER WEB-PAGE:
http://www.science.unitn.it/~priami/
[!] January
16, 2002; 11:00 hrs, [!] room 392 (Sala Conferenze IRST)
Marco Annaratone (Net-Partners) will talk about Il Venture Capital.
Abstract:Vengono presentati in modo informale le tematiche del
venture capital, i suoi obiettivi, le tipiche strutture
organizzative. Il processo che porta alla decisione di
investimento e la relazione con la societa' in cui si e'
investito vengono discussi presentando alcuni casi concreti.
Le considerazioni vengono fatte con particolare attenzione
al mondo della information technology e all'interno di una
logica di investimento in societa' ad alto contenuto
tecnologico.
January 17,
2002
No seminar.
January 24,
2002; 11:00 hrs, [!] room 392 (Sala Conferenze IRST)
Matteo Villa (TXT Italia) will talk about L' approccio di TXT al
Knowledge Management.
Abstract: Dal 1989 TXT e-solutions ha sviluppato
soluzioni software complete che rispondono alle
esigenze di un ampia gamma di imprese.
I nostri prodotti sono suddivisi in tre
principali aree: Supply Chain & Customer
Management, online Content Management, Testing &
Monitoring.
Con 12 anni di esperienza nella ricerca e sviluppo e una
forte presenza internazionale TXT partecipa a molti
progetti aziendali e internazionali di ricerca. Le
finalita` principali di queste iniziative sono
rappresentate dal continuo aggiornamento dell'offerta ai
clienti, lo sviluppo di nuovi prodotti e la creazione di
nuove competenze tecnologiche.
La presentazione, tenuta dal dott. Matteo Villa
(Project Manager) in particolare tocchera`
i seguenti aspetti:
- Presentazione azienda TXT;
- Knowledge Management in TXT: progetti esterni
(progetti di KM aziendali e di ricerca passati e
presenti);
- KM all'interno di TXT (portale aziendale,
problematiche di KM dovute alla crescita dell'azienda).
January 31,
2002
No seminar.
February 7, 2002;
11:00 hrs, [!] room 392 (Sala Conferenze IRST)
Giuliano Armano (DIEE, Universita` di Cagliari) will talk
about The Guarded-Experts Framework.
Abstract:
In this seminar, a novel framework for soft computing
-i.e., guarded experts- is presented. It entails utilizing
a population of experts, each embodying two
components: a guard and a local classifier (or a
predictor, depending on the application task). The
latter is devoted to perform classification (or
prediction) on the subset of inputs acknowledged by
the former. Given a population of guarded experts,
suitable mechanisms for performing experts selection,
fitness updating, and outputs blending must be
defined and adopted, according to the generic
constraints imposed by online systems (i.e., learning
systems, in the reinforcement learning terminology). In
this way, the proposed framework can be suitably
specialized into actual systems. Experiments have
been performed on financial time series forecasting
by utilizing a population of hybrid, genetic-neural,
experts. In fact, guards are implemented by XCS
classifiers, whereas local predictors are implemented
by artificial neural networks. Information retrieved from
technical analysis is supplied as input to guards,
whereas past stock market prices -together with other
relevant data- are used as input to predictors. The
resulting system has been tested on several stock
market indexes (i.e., S&P500, NASDAQ, and
COMIT), also taking into account trading
commissions. The results point to the good forecasting
capability of the proposed approach, which allowed
outperforming the buy-and-hold strategy, as well as
predictions obtained using locally-recurrent artificial
neural networks.
SPEAKER WEB-PAGE:
http://www.diee.unica.it/~armano/
February
14, 2002; 11:00 hrs, SRA Seminars room
Fausto Giunchiglia (DIT, Universita` di Trento and ITC-IRST) will talk
about The local relational model (or: how to apply the local model
semantics to the formalization of P2P DB integration).
Abstract: In this talk I'll provide the intuitions underlying
the use of context-related ideas to the formalization of the
integration of DB's in a P2P Environment. The formalization is based
on an evolution of Giunchiglia and Ghidini's Local models semantics
(AIJ 2001). The talk will concentrate on intuitions rather than on
formalities.
[Joint work with: Luciano Serafini, John Mylopolous, Phil Bernstein.]
SPEAKER WEB-PAGE:
http://dit.unitn.it/~fausto/
February 21, 2002;
11:00 hrs, SRA Seminars room
Luciano Serafini (ITC-IRST) will talk about Distributed Description
Logics: First results - Part 1.
Abstract: Information integration has been and remains one of the
major challenges of information processing. Description
logics can capture the content of information sources as
well as their inter-connections, and have been shown to
provide useful tools to support integration. We illustrate
the need for a more refined approach in those cases where
the original sources form a loosely federated information
system, with each one wishing to maintain its own
independent view of the world. We formalize this using the
notion of Distributed Description Logics, and provide some
initial results which explore their properties.
[Joint work with: Alexander Borgida.]
RELEVANT PAPERS:
- "Distributed First Order Logics," by C. Ghidini and L. Serafini;
http://sra.itc.it/people/serafini/distribution/frocos-98.ps.gz
- "The Local Relational Model: Model and Proof Theory," by
L. Serafini, F. Giunchiglia, J. Mylopoulos, P.A. Bernstein;
http://sra.itc.it/people/serafini/distribution/pods-02.pdf
- "Local Model Semantics, or Contextual Reasoning = Locality + Compatibility,"
by C. Ghidini and F. Giunchiglia;
http://sra.itc.it/tr/GG97b.ps.gz
SPEAKER WEB-PAGE:
http://sra.itc.it/people/serafini/
February
28, 2002; [!] 11:30 hrs, SRA Seminars room
Luciano Serafini (ITC-IRST) will talk about Distributed Description
Logics: First results - Part 2.
Abstract: This is Part 2 of the SRA Thursday seminar given on
February 21. While Part 1 was dedicated to preliminary
concepts and notions, most of them on Description Logics,
this second part will enter into the specific framework and
the "first results". Anyone with some elementary background in
Description Logics can appreciate this seminar, even if not
present to the first part.
March 7, 2002;
[!] 11:30 hrs, SRA Seminars room
Anna Perini (ITC-IRST) and Adolfo Villafiorita (ITC-IRST).
Double Talk Topic: Object Oriented vs. Agent Oriented
Methodologies for Analysis and Design.
PRESENTATION (slides):
Anna Perini:
[Download PPT, 1405K];
Adolfo Villafiorita:
[Download PPT, 203K]
Anna Perini WEB-PAGE:
http://sra.itc.it/people/perini/
Adolfo Villafiorita WEB-PAGE:
http://sra.itc.it/people/adolfo/
[!] March 11, 2002;
[!] 10:00 hrs, [!] room 392 (Sala Conferenze IRST)
Demetrio Zenti (Universita` di Trento) will present his
undergraduate thesis on Un sistema P2P per l'interrogazione di
basi di dati distribuite.
Advisor: Prof. Fausto Giunchiglia.
Co-advisor: Dr. Luciano Serafini
Abstract:
Nel campo dell'integrazione delle basi di dati stanno
emergendo delle nuove esigenze: autonomia delle basi di dati,
decentralizzazione delle informazioni, trasparenza all'utente, e
flessibilita' del sistema. Gli approcci esistenti (Global Schema
Integration, Federated Database, Multidatabase Language) non sono
in grado di soddisfare contemporaneamente queste richieste, per cui
risulta necessario cercare la soluzione altrove. Sta comparendo un
paradigma nuovo per la condivisione di informazioni, basato su
"diritti uguali" per tutti i partecipanti al sistema di condivisione:
il Peer-to-Peer (P2P). Poiche' non esiste un sistema analogo nel campo
delle basi di dati, questa tesi cerca di costruire un sistema P2P per
l'interrogazione di basi di dati distribuite.
Sfruttando il Local Relational Model (LRM) e' possibile caratterizzare
a livello formale una rete P2P di basi di dati, e dare una definizione
di interrogazione globale in un sistema siffatto. In base a questo
modello viene determinata un'architettura globale possibile del
sistema. Il cuore della tesi e' la caratterizzazione delle
corrispondenze tra gli oggetti (attributi e relazioni) dei diversi DB,
per arrivare a definire un algoritmo per la traduzione e propagazione
delle interrogazioni.
Il sistema descritto viene infine testato su tre DB per valutarne i
pregi e i limiti.
NB!
By historical tradition dating back to the early 1990s, students may
find here useful feedback and guidance to improve their work and
presentation skills. All senior researchers are warmly welcome to
attend Demetrio's talk. The talk will be given in Italian.
March 14, 2002
No seminar. (Presso la Sala Conferenze ITC-IRST viene presentato il
Contratto Nazionale della Ricerca 1998-2001, dalle 10.00 alle 12.00).
March 21, 2002;
11:00 hrs, SRA Seminars room
Paolo Bresciani (ITC-IRST) will talk about
A knowledge-based visual query system for biological databases.
Abstract:
An important task faced by bioinformatics is data and scientific
knowledge management. This activity involves several aspects, like,
for example, storing, extracting, organizing, analyzing, and make data
and knowledge assets easily usable and accessible. In particular, the
availability of intuitive interfaces that really may help the user
represents a necessary condition to gain user's acceptance and
confidence in the provided data and knowledge resource.
We tried to tackle this problem by means of a first prototypical
realization of a knowledge-based flexible tool for query formulation
against relational and object-oriented databases. Its most
interesting features are based on its capability of reasoning on the
semantics of the queries. These features are aimed at improving the
interaction between inexpert users and complex databases, thanks to
advanced knowledge reasoning algorithms. In particular, it provides a
simple, flexible, and intuitive visual interface that allow the user
to:
- interactively and iteratively build queries, exploring the semantic
of database schemas via refinement and generalization of terms;
- be prevented from building semantically inconsistent queries;
- be gradually introduced only to those part of the conceptual model
relevant for the query formulation;
- be provided with simple, but effective, features for query
refinement and query generalization.
In the present seminar we will try to demonstrate these concepts by
presenting an ongoing prototypical implementation of the system
applied to one case of biological databases, the Muscle TRAIT (ref.
muscle.cribi.unipd.it) maintained by C.R.I.B.I. (Interdepartmental
Research Center of Biotechnology) of Padua University, supported by
Telethon Foundation.
Future development include the customization of the system for the
database of the genoma of Vitis vinifera, as part of the activities
of the project "GENOMA VITE", of IASMA.
[Joint work with: Paolo Fontana (ITC-IRST).]
PRESENTATION (slides):
[Download PDF, 2791K]
SPEAKER WEB-PAGE:
http://sra.itc.it/people.epl?name=Paolo+Bresciani
March28,
2002
No seminar.
April 4, 2002;
11:00 hrs, SRA Seminars room
Marco Pistore (ITC-IRST) will talk about NuSMV2: an OpenSource tool
for symbolic model checking.
[Postponed to Thursday May 2, 2002.]
April 11, 2002;
11:00 hrs, SRA Seminars room
Paolo Traverso (ITC-IRST) will talk about Automated Software
cEngineering: 5 Steps towards a Dream.
[Postponed. New date to be announced.]
April 18, 2002;
11:00 hrs, SRA Seminars room
Paolo Busetta (ITC-IRST) will talk about Pervasive
agents.
Abstract: This seminar presents a vision driving current and future research,
rather than results of past research. Therefore, it will be very
informal and kept at an intuitive level.
Pervasive, or ubiquitous, computing is quickly becoming one of
the hottest buzzword in both research and engineering. Hardware,
driven from the continuous improvement in traditional computing
and communication equipment as well as new and astonishing
developments in the so-called nanotechnologies, is apparently
disappearing from view to be embedded everywhere, from the
clothes we wear to any sort of everyday object that sorrounds
us. An important part of this picture are some communication
technologies, such as powerlines and wireless links, which promise
to connect everything with everything else.
Integration is the great challenge to be faced in this scenario.
The issue is not simply connectivity - i.e. defining the required
interfaces - but taking advantage from, and dealing with, a large
and continuously changing number of sensors and actuators that
come in contact with each other and with humans while they move
in an environment.
We envision that most devices will be equippend with some sort
of ``intelligence'', varying from simple adaptation rules
to sophisticated recognition capabilities and goal-driven
behaviour, thus they can be called ``agents'' without abusing
the latter notion. In this scenario, integration is not by the
usual composition of elementary components in a hierarchy, but
rather by the formation of estemporary organizations of agents
with complementary capabilities.
Within the PEACH project, we are building a communication
infrastructure whose goal is to tackle some of the problems that
arise in the vision outlined above. We will describe the underlying
idea, based on group communication and overhearing; we will discuss
some of the principles behind the architecture and engineering process
being defined; finally, we will present the current state of the
project and some expected future developments.
[Joint work with: Massimo Zancanaro (ITC-IRST).]
RELEVANT PAPERS:
- "Extending Multi-Agent Cooperation by Overhearing," by
P. Busetta, L. Serafini, D. Singh, and F. Zini. IRST Technical Report
0101-01, January 2001 http://sra.itc.it/tr/BSSZ01.ps.gz.
Also in: Proceedings of the 9th International Conference on Cooperative
Information Systems (CoopIS 2001), volume 2172 of Lecture Notes in
Computer Science, Trento, Italy, September 2001. Springer Verlag.
- "Channeled Multicast for Group Communications," by Paolo Busetta,
Antonia Dona', and Michele Nori. IRST Technical Report 0111-21,
November 2001,
http://sra.itc.it/tr/BDN01.ps.gz.
To appear in: Proceedings of the 1st International Joint Conference
on Autonomous Agents & Multiagent Systems (AAMAS02), Bologna,
Italy, July 15-19, 2002.
- "Dagstuhl Seminar on Ubiquitous Computing." On-line collection of
papers and presentations,
http://www.inf.ethz.ch/vs/events/dag2001/
SPEAKER WEB-PAGE:
http://sra.itc.it/people/busetta/
April 25,
2002
No seminar: Festa della Liberazione.
May 2, 2002;
11:00 hrs, SRA Seminars room
Marco Pistore (ITC-IRST) will talk about NuSMV2: an OpenSource tool
for symbolic model checking.
Abstract: NuSMV is a symbolic model checker originated from the
reengineering,
reimplementation and extension of SMV, the first BDD-based model
checker developed at CMU. The NuSMV project aims at the development
of a state-of-the-art symbolic model checker, designed to be
applicable in technology transfer projects, well structured, open,
flexible and documented. One of the most interesting feature of NuSMV
is that it integrates the "classical" BDD-based model checking
techniques with model checking techniques based on propositional
satisfiability (SAT), that are currently enjoying a substantial
success in several industrial fields.
Starting from version 2.0 of NuSMV, we are adopting an OpenSource
development and license model. NuSMV is distributed under the LGPL
license, that allows anyone interested to freely use the tool and to
participate in its development. The aim of the NuSMV OpenSource
project is to provide to the model checking community a common
platform for the research, the implementation, and the comparison of
new symbolic model checking techniques.
In this seminar, I will present the main features of the NuSMV model
checker. Moreover, I will describe our experience in OpenSourcing the
development of NuSMV.
RELEVANT PAPERS AND LINKS:
- Papers in: http://nusmv.irst.itc.it/NuSMV/papers.html;
- Home page NuSMV: http://nusmv.irst.itc.it/.
SPEAKER WEB-PAGE:
http://sra.itc.it/people/pistore/
May 9, 2002;
11:00 hrs, SRA Seminars room
Piergiorgio Bertoli (ITC-IRST) will talk about Solving Power Supply
Restoration Problems with Planning via Symbolic Model Checking.
Abstract:
The past few years have seen a flurry of new approaches for planning
under uncertainty, but their applicability to real-world problems is
yet to be established since they have been tested only on toy
benchmark problems. To fill this gap, the challenge of solving power
supply restoration problems with existing planning tools has
recently been issued. This requires the ability to deal with
incompletely specified initial conditions, fault conditions,
unpredictable action effects, and partial observability in
real-time. This paper reports a first response to this nontrivial
challenge, using the approach of planning via symbolic
model-checking as implemented in the MBP planner. We show how the
problem can be encoded in MBP's input language, and report very
promising experimental results on a number of significant test
cases.
SPEAKER WEB-PAGE:
http://sra.itc.it/people/bertoli/
[!] May 16,
2002; 11:00 hrs, [!] room 392 (Sala Conferenze IRST)
Rosella Gennari (ILLC, Universiteit van Amsterdam) will talk about
Constraint satisfaction via propagation.
[Brought backward to Monday May 13, 2002. Please
consult the SRA Seminars
page for details on this seminar.]
RELEVANT PAPERS:
- "General Properties and Termination Conditions for Soft Constraint
Propagation," by S. Bistarelli, R. Gennari and F. Rossi;
http://staff.science.uva.nl/~rgennari/special.ps
Accepted for publication in the special issue of the Constraints
Journal (Kluwer) on soft constraints, 2001.
SPEAKER WEB-PAGE:
http://staff.science.uva.nl/~rgennari/
May 23, 2002;
11:00 hrs, SRA Seminars room
Floriano Zini (ITC-IRST) will talk about Economy-Based Optimisation
of File Access and Replication on a Data Grid.
Abstract:
We have been working on a system for the optimised access and
replication of data on a Data Grid. Our approach is based on the use of
an economic model that includes the actors and the resources in the
Grid. Optimisation is obtained via interaction of the actors in the
model, whose goals are maximising the profits of data resource
management. In the system, local optimisation results in global
optimisation through emergent marketplace behaviour. This talk gives an
overview of our model, describes the auction-based protocol we use to
select data replicas, and presents the economic reasoning required to
support optimised replication strategies.
[Joint work with: Mark Carman, Luigi Capozza and Luciano Serafini
(ITC-IRST); Kurt Stockinger (CERN, Ginevra).]
SPEAKER WEB-PAGE:
http://sra.itc.it/people/zini/
May 30, 2002;
11:00 hrs, SRA Seminars room
Roberta Cuel (PhD student, Universita` di Udine and DISA, Trento) will
talk about Knowledge Nodes: the building blocks of distributed knowledge
management.
[Postponed. New date to be announced.]
June 6, 2002
No double talk: the Forth
MRG/SRA Symposium takes place in Levanto on June 5-7.
June 13, 2002;
11:00 hrs, [!] room 392 (Sala Conferenze IRST)
Francesco Ricci (ITC-IRST) will talk about Travel Advisory Systems.
Abstract: There is a continuously growing number of web sites that support a
traveler in the selection of a travel destination or service
(e.g. flight or hotel). Actually, planning a travel towards a tourism
destination is a complex problem solving activity. Therefore, major
eCommerce web sites dedicated to travel and tourism have recently
started to better cope with leisure travel planning, incorporating
recommender systems, i.e. applications that provide advice to users
about products they might be interested in. Recommender systems for
travel planning try to mimic the interactivity observed in traditional
counselling sessions with travel agents.
In this talk I'll present a web based recommender system aimed at
supporting a user in information filtering and product bundling. The
system enables the selection of travel locations, activities and
attractions, and supports the bundling of a personalized travel bag. A
travel bag is composed in a mixed initiative way: the user poses queries
and the recommender exploits an innovative technology that helps the
user, when needed, to reformulate the query. Travel bags are stored in a
memory of cases, which is exploited for ranking travel items extracted
from catalogues. A new 'collaborative' approach is introduced, where
user past behaviour similarity is replaced with session (travel bag)
similarity.
PRESENTATION (slides):
[Download PDF, 2132K]
RELEVANT PAPERS:
- "Intelligent query managment in a mediator architecture," by
Francesco Ricci, Nader Mirzadeh, Adriano Venturini;
[PDF, 272K].
To appear in: Proceedings of the 1st IEEE International Symposium "Intelligent
Systems", Vladimir Jotsov (ed.), 2002.
- "ITR: a case-based travel advisory system," by F. Ricci,
B. Arslan, N. Mirzadeh, A. Venturini;
[PDF, 623K]
To appear in: Proceedings of 6th European Conference on Case Based
Reasoning (ECCBR), S. Craw (ed.), Springer Verlag, 2002.
SPEAKER WEB-PAGE:
http://ectrl.itc.it/home/laboratory/people/peopleResult.jsp?lastname=Ricci&username=ricci
June 20, 2002;
11:00 hrs, SRA Seminars room
Bernardo Magnini (ITC-IRST) will talk about Using Lexical Knowledge
Bases for NLP Applications.
Abstract: In this seminar I will briefly review the use of
Lexical Knowledge
Bases in Natural Language Processing. I will focus on WordNet,
probably the most popular repository of word meanings. I will
introduce the basic relations in WordNet and I will give an overview
of current attempts aiming at enriching WordNet with new information,
including multilingual versions of WordNet.
Then, I will report recent approaches that use WordNet to address
content-based processing of open-domain text collections. I will
mention both successes and drawbacks in Word Sense Disambiguation,
Question/Answering, query expansion for Information Retrieval and
Conceptual Indexing. Finally, I will talk about the use of WordNet for
Meaning Negotiation, a new scenario recently proposed in the context
of the Edamok project.
RELEVANT PAPERS AT:
- http://tcc.itc.it/research/textec/topics/multiwordnet/english/pub-frame.html
- http://tcc.itc.it/research/textec/topics/disambiguation/index.html
SPEAKER WEB-PAGE:
http://tcc.itc.it/people/magnini.html
June 27, 2002;
11:00 hrs, SRA Seminars room [DIT]
Peter H. Schmitt (Karlsruhe University, Germany) will talk about The KeY
Project: Formal Methods in the Software Engineering Process.
Abstract: Research in the field of Formal Methods, and in particular
in the area of program verification by mathematical proofs, has
produced surprising progress during the last five years.
Practical feasability has been documented by numerous case studies
on real world problems. Nevertheless, the use of formal methods is
still for the greatest part restricted to academic projects.
We believe that one of the main obstacle to wide spread use is the
lack of integration into the processes, methods and tools used
in everyday software development.
The KeY project, a joint effort by the University of Karlsruhe, Germany,
and Chalmers University, Goteborg, Sweden, aims at improving this
situation.
I will start with a short overview of the KeY system and then
zoom in on two topics in greater detail:
(1) The extension of Dynamic Logic by non-static function symbols
and the constructs refering to values before program execution.
(2) The translation of UML/OCL specifications into Dynamic Logic.
At the end there will be a demo (around 30') on the integration of the
techniques in the state-of-the-art UML/SwEng tool Together.
SPEAKER WEB-PAGE:
http://i12www.ira.uka.de/~pschmitt/
The next schedule is available at
this
location. Send suggestions for speakers and events to
the current organizer.
[
Home |
Links |
Local |
People |
Provers&Tools |
Publications |
Research |
Applications |
Teaching ]
Automated Reasoning Systems Division (SRA)
Contact Address
This page was last built for SRA by Alessandro Agostini
on Wed, Dec 05, 2001 at 16:04:00 MET.
|
| |
|
© 2002 SRA, ITC-IRST |
|