Note: This bibliographic page is archived and will no longer be updated. For an up-to-date list of publications from the Music Technology Group see the Publications list .

Computational Modeling of Statistical Learning of Tone Sequences (Poster)

Title Computational Modeling of Statistical Learning of Tone Sequences (Poster)
Publication Type Conference Paper
Year of Publication 2007
Conference Name International Conference on Cognitive and Neural Systems (ICCNS 2007)
Authors Hazan, A. , Herrera P. , Marxer R. , Grachten M. , & Purwins H.
Abstract In this work, we are interested in building computational models that are able to learn and represent the statistical regularities of tone sequences by operating in a way that is constrained by findings in cognitive science. We consider recurrent architectures (here Simple Recurrent Networks [Elman90, Cleemermans91]) and chunking models (Feedforward neural networks) to model this task in an on-line setting. This means that the model's internal state is updated on the fly [Kuhn06] instead of running through the sequence several times (batch learning). We aim to simulate the findings obtained by [Saffran99], in which human subjects are able to acquire the statistical regularities of tone sequences. We propose a validation loop that follows the experimental setup that was used with human subjects in order to characterize the models accuracy to learn the statistical regularities of tone sequence as observed by [Saffran99]