Robust Classification and Change Detection for Brain-Computer Interfaces

ABSTRACT: In this talk, we will first discuss eye movement decoding in a working memory experiment involving a macaque monkey. Our objective is to use the local field potentials (LFPs) collected from the brain of the monkey to decode the type of task that the monkey is doing, and the direction of saccade in each task. We will show that the LFP time-series data can be modeled using a nonparametric regression framework, and show that the classifiers trained using minimax function estimators as features are robust and consistent. We will also discuss application of the resulting classifier to the brain data. 

We will then briefly discuss the problem of change detection apply it to spike data from a mice experiment collected using cues and electric shocks.

This is a joint work with Taposh Banerjee.

BIO: Vahid Tarokh is Rhodes Family Professor of Electrical and Computer Engineering, Professor of Mathematics, and Computer Science at Duke University.

He worked at AT&T Labs-Research until 2000, and subsequently at MIT (as an Associate Professor of EECS) until 2002. He joined Harvard University as Perkins Professor of Applied Mathematics and Hammond Vinton Hayes Senior Fellow of Electrical Engineering. He then joined Duke University in January 2018. His current research focuses on statistical signal processing and applications. Dr. Tarokh has received a number of awards, and holds four honorary degrees.

Hosted by: Paul Bogdan