Workshop: Music, Mathematics and Computation

This workshop intends to establish interdisciplinary contacts within the fields of music, mathematics and computation, and to get an overview of the research that is done in some prominent research groups. Furthermore, we hope to start a discussion about the research methods in these fields and to reach a consensus about the importance of the combination of these research disciplines. Everybody is cordially invited to participate; attendance is free. However, for reasons of planning the room, please send me an e-mail if you want to participate.
The workshop is subsidized by the Evert Willem Beth Stichting.

The workshop Music, Mathematics and Computation is collocated with the PhD defense ceremony of Aline Honingh, that will take place on Friday October 20th. Please feel free to attend the ceremony and to ask for a free copy of my dissertation, entitled The Origin and Well-Formedness of Tonal Pitch Structures.
Date: Saturday October 21st 2006
Place: Institute for Logic, Language and Computation, University of Amsterdam, Plantage Muidergracht 24, room P.017
Organizer: Aline Honingh


Schedule

13:00 - 13:30 Rens Bod, University of St. Andrews/ University of Amsterdam
13:30 - 14:00 Elaine Chew, University of Southern California
14:00 - 14:30 Thomas Noll, Univeristy of Berlin/ ESMUC, Barcelona
14:30 - 15:00 coffee break
15:00 - 15:30 Henkjan Honing, University of Amsterdam
15:30 - 16:00 Frans Wiering and Anja Volk, Utrecht University
16:00 - 17:00 Discussion session chaired by Aline Honingh

Abstracts

Rens Bod, University of St. Andrews/ University of Amsterdam
Title: Simplicity vs Likelihood in Computational Musicology
Abstract: The controversy between simplicity-based and likelihood-based approaches has a long history in Artificial Intelligence and Cognitive Science. In this talk I will review some simplicity and likelihood measures and show how they may be put to use in bootstrapping melodic structure. I will argue that the notion of Most Compact Structure, as worked out by Aline Honingh, may be viewed as a measure of simplicity, which is one of two directions of the NWO research program ``Unifying Models of Linguistic, Musical and Visual Processing".

Elaine Chew, University of Southern California
Title: Building an interdisciplinary research program in computational music cognition
Abstract: This presentation will give an overview of the research conducted at the Music Computation and Cognition Laboratory at the University of Southern California Viterbi School of Engineering. The narration will be somewhat chronological and thematic, showing the development of the research group and ideas over the past five years. Featured in the talk will be projects ranging from computational music analysis and visualization to expressive performance analysis and synthesis, the cultivation and recruitment of students for research in the area, and the development of a topical course on engineering approaches to music cognition.

Thomas Noll, Univeristy of Berlin/ ESMUC, Barcelona
Title: Musical Intervals and Special Linear Transformations
Abstract: For the abstract and paper, please see http://flp.cs.tu-berlin.de/~noll/MIST.pdf

Henkjan Honing, University of Amsterdam
Title: Computational modeling of music cognition: The role of surprise in model selection
Abstract: While the most common way of evaluating a computational model is by showing a good fit with the empirical data, recently the literature on theory testing and model selection criticizes the assumption that this is actually strong evidence for a model. This paper explores the possibilities of developing a method selection technique that can serve as an alternative to a goodness-of-fit (GOF) measure. This alternative, a measure of surprise, is based on the common idea that a model gets more support from the correct prediction of an unlikely event than the correct prediction of something that was expected anyway.

Frans Wiering and Anja Volk, Utrecht University
Title: Creating a Search Engine for Folksongs... and Other Music
Abstract: The Meertens Instituut of the KNAW hosts a large collection of Dutch folksongs, named "Onder de groene linde", consisting of some 7000 field recordings and 4000 transcriptions. In the recently-started WITCHCRAFT project, we design and test a melody search engine that will provide better access to both end users and researchers of this collection. The principal tasks for the engine to perform are:
  • identifying a given melody: often one melody can have several different texts
  • finding variants of a melody: melodies change by being orally transmitted
For the latter task, we investigate mental models of melody. The current demo version of the system, which will be demonstrated in this talk, does not yet incorporate this research, but we firmly believe that music searching can be improved a lot by taking such models into account.


Aline Honingh
Last modified: Thu Oct 12 17:52:27 CEST 2006