Dear list members, Please find below our final call for submissions to our special collection of Music & Science on “Explaining music with AI: Advancing the scientific understanding of music through computation”. Note that there is no need to send an _expression_ of interest at this stage. Note also that the deadline for submission is 31 August 2022 and that we expect to have the collection published by May 2023. Kind regards, David Meredith, Anja Volk and Tom Collins Final Call for Papers for a Special Collection of Music & Science on “Explaining music with AI: Advancing the scientific understanding of music through computation” Guest edited by David Meredith, Anja Volk & Tom Collins In recent years, a huge number of publications, particularly in the areas of music information retrieval and music generation, have reported on projects in which deep learning neural network models have been used successfully to carry out a wide variety of generation, regression and classification tasks on musical data. This work has significantly contributed to the arsenal of computational tools that we have at our disposal, if we want to explore, organise, create or simply enjoy digital music resources. However, the majority of such models are “black box” models that have thousands or even millions of free parameters, whose values are determined through training on, typically, large amounts of data. The computing pioneer, John von Neumann, allegedly joked, “With four free variables I can fit an elephant and with five I can make him wiggle his trunk” (Freeman Dyson, 2004, “A meeting with Enrico Fermi”, Nature, 427:297). Such considerations prompt us to question whether such black-box deep learning models make a significant contribution to our scientific understanding of music, musical processes and musical behaviour. For this special collection, we seek high quality contributions that report on recent research in which any computational method has been used to advance our understanding of how and why music is created, communicated and received. We are particularly interested in shining a light on computational methods that have perhaps not received the attention they deserve because of the dominance of deep learning in recent years. At the same time, contributions in which deep learning and other neural network models have been shown to advance the scientific understanding of music are also very welcome. Submissions may address any aspect of musical behaviour, including, but not limited to, composition, improvisation, performance, listening, musical gestures and dance. Contributions may also focus on any aspects of music, e.g., rhythm, harmony, melody, counterpoint, instrumentation or timbre. We likewise set no constraints on the considered music’s style, genre, period or place of origin. However, the reported work must have adopted a computational approach that has led to an advancement in our scientific understanding of music. We are also keen to cover a variety of application areas where music is put to use, not only for pure entertainment or artistic purposes, but also, for example, in healthcare, in rituals and ceremonies, in meditation, in film soundtracks or video games, or in politics or advertising. We welcome, in particular, contributions where a computational approach has been employed in conjunction with methodologies and knowledge from other fields, such as psychology, neuroscience, musicology, sociology, biology, physics, anthropology or ethnomusicology. The deadline for submission of full manuscripts is Wednesday 31 August 2022. Full manuscripts will need to be submitted using the online submission system at https://mc.manuscriptcentral.com/mns and should follow the Submission Guidelines, which can be accessed at https://journals.sagepub.com/author-instructions/MNS. Accepted papers in this collection will be published with open access by the end of May 2023. Important dates
Kind regards, David Meredith Anja Volk Tom Collins |