FastICA blind source separation for microscopy images with BM3D denoising and ICASSO robustness analysis

Fast implementations of FastICA and DUET for blind source separation

A class research project on independent component analysis (ICA) and FastICA for Matrix Analysis for Signal Processing and Machine Learning (ECE 599) course at Oregon State University.

C implementation of FastICA algorithm

Implementation of FastICA algorithm for PULP platform

Code for a CPU, GPU and a Hybrid version of the FastICA algorithm.

For university purposes only!

An implementation of the fastICA algorithm for the cocktail party problem

Project used to split N mixed audio files into N separated independent audio files. This project uses FastICA algorthm.

Explore machine learning for automotive testing optimization. Predictive analytics to reduce testing time and environmental impact.

Прогнозирование температуры звезд с применением нейронных сетей.

Code for 'Estimating Treatment Effects with Independent Component Analysis' (arXiv:2507.16467)