Fast implementations of FastICA and DUET for blind source separation

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

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

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

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

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

Blind source separation system in MATLAB using PCA and ICA to recover audio sources from noisy linear mixtures. Compares methods using correlation, kurtosis, and signal reconstruction quality, showing ICA superior for independent signal recovery.

Прогнозирование температуры звезд с применением нейронных сетей.