# Information Theoretic Feature Transformation Learning for Brain Interfaces

## TL;DR

This paper introduces information theoretic feature transformation methods, MMI-LinT and MMI-NonLinT, which improve brain-computer interface decoding performance over traditional feature selection techniques by maximizing mutual information.

## Contribution

It proposes novel mutual information-based linear and nonlinear feature transformation frameworks for brain interfaces, extending beyond conventional feature selection methods.

## Key findings

- Significantly improved decoding accuracy with proposed methods
- Effective in both binary and multi-class BCI tasks
- Outperforms traditional feature selection techniques

## Abstract

Objective: A variety of pattern analysis techniques for model training in brain interfaces exploit neural feature dimensionality reduction based on feature ranking and selection heuristics. In the light of broad evidence demonstrating the potential sub-optimality of ranking based feature selection by any criterion, we propose to extend this focus with an information theoretic learning driven feature transformation concept. Methods: We present a maximum mutual information linear transformation (MMI-LinT), and a nonlinear transformation (MMI-NonLinT) framework derived by a general definition of the feature transformation learning problem. Empirical assessments are performed based on electroencephalographic (EEG) data recorded during a four class motor imagery brain-computer interface (BCI) task. Exploiting state-of-the-art methods for initial feature vector construction, we compare the proposed approaches with conventional feature selection based dimensionality reduction techniques which are widely used in brain interfaces. Furthermore, for the multi-class problem, we present and exploit a hierarchical graphical model based BCI decoding system. Results: Both binary and multi-class decoding analyses demonstrate significantly better performances with the proposed methods. Conclusion: Information theoretic feature transformations are capable of tackling potential confounders of conventional approaches in various settings. Significance: We argue that this concept provides significant insights to extend the focus on feature selection heuristics to a broader definition of feature transformation learning in brain interfaces.