RT Journal Article
SR Electronic
T1 Machine learning for neural decoding
JF eneuro
JO eNeuro
FD Society for Neuroscience
SP ENEURO.0506-19.2020
DO 10.1523/ENEURO.0506-19.2020
A1 Joshua I. Glaser
A1 Ari S. Benjamin
A1 Raeed H. Chowdhury
A1 Matthew G. Perich
A1 Lee E. Miller
A1 Konrad P. Kording
YR 2020
UL http://www.eneuro.org/content/early/2020/07/31/ENEURO.0506-19.2020.abstract
AB Despite rapid advances in machine learning tools, the majority of neural decoding approaches still use traditional methods. Modern machine learning tools, which are versatile and easy to use, have the potential to significantly improve decoding performance. This tutorial describes how to effectively apply these algorithms for typical decoding problems. We provide descriptions, best practices, and code for applying common machine learning methods, including neural networks and gradient boosting. We also provide detailed comparisons of the performance of various methods at the task of decoding spiking activity in motor cortex, somatosensory cortex, and hippocampus. Modern methods, particularly neural networks and ensembles, significantly outperform traditional approaches, such as Wiener and Kalman filters. Improving the performance of neural decoding algorithms allows neuroscientists to better understand the information contained in a neural population and can help advance engineering applications such as brain machine interfaces.Significance statement Neural decoding is an important tool for understanding how neural activity relates to the outside world and for engineering applications such as brain computer interfaces. Despite many advances in machine learning, it is still common to use traditional linear methods for decoding. Here, we present a tutorial and accompanying code package so that neuroscientists can more easily implement machine learning tools for neural decoding.