Exploring the trade-off between deep-learning and explainable models for brain-machine interfaces Luis H. Cubillos 1, Matthew J. Mender 1, Joseph T. Costello