# Faster ICA under orthogonal constraint

arXiv: https://arxiv.org/abs/1711.10873 · HTML: https://mlsanity.com/paper/1711.10873/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Nov 2017.

## Built on by (most cited first, 9 of 9)

- [Fast and accurate optimization on the orthogonal manifold without retraction](https://mlsanity.com/paper/2102.07432/) (2021)
- [Infeasible Deterministic, Stochastic, and Variance-Reduction Algorithms for Optimization under Orthogonality Constraints](https://mlsanity.com/paper/2303.16510/) (2023)
- [Multi-View Independent Component Analysis with Shared and Individual Sources](https://mlsanity.com/paper/2210.02083/) (2022)
- [Optimization without Retraction on the Random Generalized Stiefel Manifold](https://mlsanity.com/paper/2405.01702/) (2024)
- [Geometric Neural Network based on Phase Space for BCI-EEG decoding](https://mlsanity.com/paper/2403.05645/) (2024)
- [Alljoined1 -- A dataset for EEG-to-Image decoding](https://mlsanity.com/paper/2404.05553/) (2024)
- [OptGAN: Optimizing and Interpreting the Latent Space of the Conditional Text-to-Image GANs](https://mlsanity.com/paper/2202.12929/) (2022)
- [RRCNN: A novel signal decomposition approach based on recurrent residue convolutional neural network](https://mlsanity.com/paper/2307.01725/) (2023)
- [Beyond Correlation: Learning Supervised, Sample-Distinct, and Eigenimage-Interpretable Representations](https://mlsanity.com/paper/2507.21136/) (2025)

## Abstract (arXiv)

Independent Component Analysis (ICA) is a technique for unsupervised exploration of multi-channel data widely used in observational sciences. In its classical form, ICA relies on modeling the data as a linear mixture of non-Gaussian independent sources. The problem can be seen as a likelihood maximization problem. We introduce Picard-O, a preconditioned L-BFGS strategy over the set of orthogonal matrices, which can quickly separate both super- and sub-Gaussian signals. It returns the same set of sources as the widely used FastICA algorithm. Through numerical experiments, we show that our method is faster and more robust than FastICA on real data.