Model evaluation plays a special role in interactive machine learning (IML) systems in which users rely on their assessment of a model's performance in order to determine how to improve it. A better understanding of what model criteria are important to users can therefore inform the design of user interfaces for model evaluation as well as the choice and design of learning algorithms. We present work studying the evaluation practices of end users interactively building supervised learning systems for real-world gesture analysis problems. We examine users' model evaluation criteria, which span conventionally relevant criteria such as accuracy and cost, as well as novel criteria such as unexpectedness. We observed that users employed evaluation techniquesincluding cross-validation and direct, real-time evaluationnot only to make relevant judgments of algorithms' performance and interactively improve the trained models, but also to learn to provide more effective training data. Furthermore, we observed that evaluation taught users about what types of models were easy or possible to build, and users sometimes used this information to modify the learning problem definition or their plans for using the trained models in practice. We discuss the implications of these findings with regard to the role of generalization accuracy in IML, the design of new algorithms and interfaces, and the scope of potential benefits of incorporating human interaction in the design of supervised learning systems.

In 2005, the Princeton Laptop Orchestra (PLOrk), an ensemble of fifteen laptop-based meta-instruments, began its first season. In this article, the author explores the motivations for starting a laptop orchestra, both in musical and cultural terms, and some of the aesthetic, technical, and compositional issues that face those interested in working with such an ensemble.

No abstract

No abstract

Traditional musical instruments provide compelling metaphors for human-computer interfacing, both in terms of input (physical, gestural performance activities) and output (sound diffusion). The violin, one of the most refined and expressive of traditional instruments, combines a peculiar physical interface with a rich acoustic diffuser. We have built a new instrument that includes elements of both the violin's physical performance interface and its spatial filtering audio diffuser, yet eliminates both the resonating body and the strings. The instrument, BoSSA (Bowed-Sensor-Speaker-Array), is an amalgamation and extension of our previous work with violin interfaces, physical models, and directional tonal radiation studies. In addition to describing the various physical and software elements that make up BoSSA, we discuss some of its musical features and potentials; we are particularly impressed by the sense of presence and intimacy it provides, and by its potential for creating a new kind of electronic chamber music.