October 2013 Archives

The final lecture schedule

| No Comments
As my grant is now winding down, I am in the final push to push out a few more articles, and go on an evangelism spree. Watch for when I am in your town!

  1. Wednesday Oct. 30, 18h30, OFAI, Vienna
  2. Wednesday Nov. 13, 15h30, MTG, Barcelona
  3. Monday Nov. 25, 14h00, Télécom ParisTech, Paris
  4. Tuesday Nov. 26, all day, AAU, Copenhagen
  5. Wednesday Dec. 4, 11h00 Fraunhofer IAIS, Bonn
My current show is called, "The crisis of evaluation in MIR".

Abstract: I critically address the "crisis of evaluation" in music information retrieval (MIR), with particular emphasis paid to music genre recognition, music mood recognition, and autotagging. I demonstrate four things: 1) many published results unknowingly use datasets with faults that render them meaningless; 2) state-of-the-art ("high classification accuracy") systems are fooled by irrelevant factors; 3) most published results are based upon an invalid evaluation design; and 4) a lot of work has unknowingly built, tuned, tested, compared and advertised "horses" instead of solutions. (The example of the horse Clever Hans provides an appropriate illustration.) I argue these problems occur because: 1) many researchers assume a dataset is a good dataset because many others use it; 2) many researchers assume evaluation that is standard in machine learning or information retrieval are useful and relevant for MIR; 3) many researchers mistake systematic, rigorous, and standardized evaluation for being scientific evaluation; and 4) problems and success criteria remain ill-defined, and thus evaluation poor, because researchers do not define appropriate use cases. I show how this "crisis of evaluation" can be addressed by formalizing evaluation in MIR to make clear its aims, parts, design, execution, interpretation, and assumptions. I also present several alternative evaluation approaches that can separate horses from solutions.

PhD course Nov. 25-29, 2013

| No Comments
Topics in Music Analysis, Cognition and Synthesis,

Doctoral School of Engineering and Science at Aalborg University Copenhagen

The course consists of three parts covering the topics of music analysis, cognition, and synthesis with emphasis on recent advances. The first part covers methods and models for music analysis. This includes statistical methods for parameterization of music signals and methods for music information retrieval. In the second part, models of music perception and cognition are covered, including analysis of symbolic representations of music. Finally, physical models of music instruments are covered in the last part along with parametric and interactive methods for synthesis and spatialization of sounds.

Prerequisites: Basic knowledge of sound and music computing

Blog Roll

About this Archive

This page is an archive of entries from October 2013 listed from newest to oldest.

September 2013 is the previous archive.

November 2013 is the next archive.

Find recent content on the main index or look in the archives to find all content.