A visit to the the Intelligent Music Processing and Machine Learning Group of the Austrian Research Institute for Artificial Intelligence (OFAI)

Assistant Researcher Angelos Gkiokas from the ILSP visited for three months the Intelligent Music Processing and Machine Learning Group of the Austrian Research Institute for Artificial Intelligence (OFAI).

The aim was the knowhow transfer from OFAI to ILSP in the field of Music Information Retrieval as well as establishing a closer research collaboration between institutes. More specifically, the visiting researcher collaborated with OFAI’s research teams in the following topics: a. establishing data-driven robust rhytm representations based on deep learning techniques b. evaluating mid and high level descriptions for rhythm analysis.

The impact of this visit was the strengthen of ILSP’s research potential in the following domains: Deep Learning Techniques, Unsupervised Feature Learning for Rhythm Classification.

Dissemination Activities include:

-          3.3.15: Aggelos Gkiokas’ talk at the OFAI under the title “Deploying Deep Belief Nets on the Periodicity Function for Rhythm Analysis”

-          19.03.2015: Aggelos Gkiokas’ talk at Johannes Kepler University of Linz under the  title “Deploying Deep Belief Nets on the Periodicity Function For Rhythm Analysis”

-          5.06.2015: Aggelo’s Gkiokas’ talk at OFAI under the title “Towards and Invertible Rhythm Representation”

-          Submission of the paper “Towards an Invertible  Rhytm Representation” on the 18th International Conference on Digital Audio Effects