Back
Predicting EMG Data from M1 Neurons with Variational Bayesian Least Squares
An increasing number of projects in neuroscience requires the statistical analysis of high dimensional data sets, as, for instance, in predicting behavior from neural firing, or in operating artificial devices from brain recordings in brain-machine interfaces. Linear analysis techniques remain prevalent in such cases, but classi-cal linear regression approaches are often numercially too fragile in high dimen-sions. In this paper, we address the question of whether EMG data collected from arm movements of monkeys can be faithfully reconstructed with linear ap-proaches from neural activity in primary motor cortex (M1). To achieve robust data analysis, we develop a full Bayesian approach to linear regression that automatically detects and excludes irrelevant features in the data, and regular-izes against overfitting. In comparison with ordinary least squares, stepwise re-gression, partial least squares, and a brute force combinatorial search for the most predictive input features in the data, we demonstrate that the new Bayesian method offers a superior mixture of characteristics in terms of regularization against overfitting, computational efficiency, and ease of use, demonstrating its potential as a drop-in replacement for other linear regression techniques. As neuroscientific results, our analyses demonstrate that EMG data can be well pre-dicted from M1 neurons, further opening the path for possible real-time inter-faces between brains and machines.
@inproceedings{Ting_ANIPS_2005, title = {Predicting EMG Data from M1 Neurons with Variational Bayesian Least Squares}, booktitle = {Advances in Neural Information Processing Systems 18 (NIPS 2005)}, abstract = {An increasing number of projects in neuroscience requires the statistical analysis of high dimensional data sets, as, for instance, in predicting behavior from neural firing, or in operating artificial devices from brain recordings in brain-machine interfaces. Linear analysis techniques remain prevalent in such cases, but classi-cal linear regression approaches are often numercially too fragile in high dimen-sions. In this paper, we address the question of whether EMG data collected from arm movements of monkeys can be faithfully reconstructed with linear ap-proaches from neural activity in primary motor cortex (M1). To achieve robust data analysis, we develop a full Bayesian approach to linear regression that automatically detects and excludes irrelevant features in the data, and regular-izes against overfitting. In comparison with ordinary least squares, stepwise re-gression, partial least squares, and a brute force combinatorial search for the most predictive input features in the data, we demonstrate that the new Bayesian method offers a superior mixture of characteristics in terms of regularization against overfitting, computational efficiency, and ease of use, demonstrating its potential as a drop-in replacement for other linear regression techniques. As neuroscientific results, our analyses demonstrate that EMG data can be well pre-dicted from M1 neurons, further opening the path for possible real-time inter-faces between brains and machines.}, editors = {Weiss, Y.;Schölkopf, B.;Platt, J.}, publisher = {Cambridge, MA: MIT Press}, address = {Vancouver, BC, Dec. 6-11}, year = {2005}, note = {clmc}, slug = {ting_anips_2005}, author = {Ting, J. and D'Souza, A. and Yamamoto, K. and Yoshioka, T. and Hoffman, D. and Kakei, S. and Sergio, L. and Kalaska, J. and Kawato, M. and Strick, P. and Schaal, S.}, crossref = {p2577}, url = {http://www-clmc.usc.edu/publications/T/ting-NIPS2005.pdf} }