SPARSE INFOMAX BASED ON HOYER PROJECTION AND ITS APPLICATION TO SIMULATED STRUCTURAL MRI AND SNP DATA
K. Duan, R. F. Silva, J. Chen, D. Lin, V. D. Calhoun, J. Liu
This paper introduces a sparse infomax algorithm using Hoyer projection for improved pattern recovery in simulated structural MRI and SNP data, showing better sparseness and performance under low signal-to-noise ratio conditions.
bioRxiv · Mar 7, 2019 · PreprintBioengineering