International
School "Eduardo R. Caianiello", 10th Course
and
Workshop of the PASCAL Network of
Excellence
The Analysis of Patterns
Centre "Ettore Majorana" for
Scientific Culture, Erice, Italy
October 28 - November 6, 2005
Directors of the Course:
Nello Cristianini,
University of California, Davis, USA
Raffaele Cerulli,
Universita' di Salerno, Italy
John Shawe-Taylor,
University of Southampton, UK
Director of the School "Eduardo Caianiello":
Maria
Marinaro, Universita' di Salerno, Italy
Description
Automatic pattern analysis of data is a pillar of modern science, technology and business, with deep roots in statistics, machine learning, pattern recognition, theoretical computer science, and many other fields. A unified conceptual understanding of this strategic field is of utmost importance for researchers as well as for users of this technology. This workshop-course will emphasizes the common principles and roots of modern pattern analysis technology, developed independently by many different scientific communities over the past 30 years, and their impact on modern science and technology.
Pattern detection and discovery is at the center of many disciplines, ranging from classical statistics to modern artificial intelligence, including bioinformatics, web analysis and many more. In a way, all science has become more data-driven in the last decades, and increasingly relies on mining massive datasets for any useful relations, as is for example the case in computational genomics.
The emphasis on different techniques, problems and application domains has given rise to several separate communities, parallelly and independently working on very related topics. Machine learning, data mining, statistical pattern recognition, syntactical pattern recognition, combinatorial pattern matching, and classical statistics have less interactions than one would imagine, although they all deal with the general problem of finding and analyzing relations in data. Applications (from genomics to web) which cross these traditional boundaries have recently provided new incentives to pursue a unified view.
Automatically finding trends, anomalies, similarities and any other
relation of interest in a dataset is a crucial task for theory and applications,
where statistical and algorithmic ideas are intertwined, as well as ideas
and methods from information theory, optimization, data mining and machine
learning. Very often, different communities focus on different aspects
or approaches, so that a general view of the problem is difficult to achieve.
The goal of this course is precisely this: to bring together and compare
all existing approaches to the problem of detecting and analyzing any
type of patterns in any type of data.
This course will deal with general themes arising from the analysis of
patterns in different disciplines, and will bring together international
experts dealing with the different types of pattern analysis, pattern
recognition, matching, discovery etc. etc. The intended audience are students
and researchers in statistcs, computer science, data mining, neural networks
and data intensive sciences, interested in pattern analysis. The focus
will be on unifying principles that underlie classic disciplines such
as sequence pattern matching, pattern recognition by means of machine
learning systems, etc.
Among others, these are subdisciplines that will be brought together in this meeting: combinatorial pattern matching; statistical pattern recognition; structural and syntactical pattern recognition; pattern theory; pattern formation; probabilistic graphical modeling; pattern matching in graphs; random graphs evolution; computer vision; machine learning; knowledge discovery in databases and data mining; algorithmic information theory; etc ...
Just to provide an indication of the diversity of communities which might
be interested in this meeting, we list some of the reference conferences
of these communities: ICML, COLT, UAI, NIPS, CPM, ICPR, CVPR, ALT, RECOMB,
ISMB, WWW, ASA, KDD, ....