Improved Estimation in Time Varying Models
This work addresses variance reduction in time-varying models, which is an incremental improvement for applications like brain-computer interfaces.
The paper tackles the high variance problem in locally adapted parameterizations, such as in time-varying models, by estimating a transformed space and local parameters simultaneously, resulting in improved estimation as demonstrated on synthetic data and a BCI EEG classification task.
Locally adapted parameterizations of a model (such as locally weighted regression) are expressive but often suffer from high variance. We describe an approach for reducing the variance, based on the idea of estimating simultaneously a transformed space for the model, as well as locally adapted parameterizations in this new space. We present a new problem formulation that captures this idea and illustrate it in the important context of time varying models. We develop an algorithm for learning a set of bases for approximating a time varying sparse network; each learned basis constitutes an archetypal sparse network structure. We also provide an extension for learning task-driven bases. We present empirical results on synthetic data sets, as well as on a BCI EEG classification task.