Sparse and Compressive Sensing Techniques

The paper develops a method to detect cycles in sparse, irregularly sampled neuro-behavioral timeseries using basis pursuit denoising with polynomial detrending, validated on epileptiform discharge data.

This paper applies compressive sensing to reverse correlation to reduce the number of stimulus-response pairs needed for cognitive representation reconstruction without bias.

This paper develops an algorithm for imputing missing values in non-negative matrix factorization, showing it performs as well as or better than existing methods.

This paper develops a theoretical neuroscience model for sparse coding in compressed neural representations that respects local wiring constraints in cortical communication.

This study compares image quality between total variation minimization and nonlinear sparsifying transform-based compressed sensing in sparse-view CT reconstruction.

The paper introduces MAGE, a method for detecting transient oscillatory burst amplitude and phase with improved accuracy using adaptive Gabor expansions.

The paper develops multiplicative update rules for optimizing convex objectives with nonnegativity constraints and applies them to signal processing tasks using neuronal calcium imaging and electrophysiology data.