Functional Kernel Hypothesis Testing for Channel Selection in Time Series Classification


Multi-channel time-series classification is a challenging task. With sometimes thousands of sensors available for real-wolrd applications, it is a daunting and difficult task to select which channels to include and which not.

This research aims to handle the following scenario: Given a Time Series Classification (TSC) task on time series T with M channels (sensors), find the most P (much smaller than M) relevant channels for predicting the segment-wise labels L.

See the poster