# Iterative Gaussianization: from ICA to Random Rotations

arXiv: https://arxiv.org/abs/1602.00229 · HTML: https://mlsanity.com/paper/1602.00229/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jan 2016.

## Built on by (most cited first, 26 of 26)

- [Feature Shift Detection: Localizing Which Features Have Shifted via Conditional Distribution Tests](https://mlsanity.com/paper/2107.06929/) (2021)
- [Nonparametric Generative Modeling with Conditional Sliced-Wasserstein Flows](https://mlsanity.com/paper/2305.02164/) (2023)
- [Machine learning for modelling unstructured grid data in computational physics: a review](https://mlsanity.com/paper/2502.09346/) (2025)
- [The Art of Deception: Color Visual Illusions and Diffusion Models](https://mlsanity.com/paper/2412.10122/) (2024)
- [Modeling Unknown Stochastic Dynamical System Subject to External Excitation](https://mlsanity.com/paper/2406.15747/) (2024)
- [Exploring Dimensionality Reduction Techniques in Multilingual Transformers](https://mlsanity.com/paper/2204.08415/) (2022)
- [On the Importance of Gaussianizing Representations](https://mlsanity.com/paper/2505.00685/) (2025)
- [On the Convergence Rate of Gaussianization with Random Rotations](https://mlsanity.com/paper/2306.13520/) (2023)
- [Iterative Alignment Flows](https://mlsanity.com/paper/2104.07232/) (2021)
- [Differentiable Gaussianization Layers for Inverse Problems Regularized by Deep Generative Models](https://mlsanity.com/paper/2112.03860/) (2021)
- [Unsupervised Anomaly and Change Detection with Multivariate Gaussianization](https://mlsanity.com/paper/2204.05699/) (2022)
- [Orthonormal Convolutions for the Rotation Based Iterative Gaussianization](https://mlsanity.com/paper/2206.03860/) (2022)
- [Image Statistics Predict the Sensitivity of Perceptual Quality Metrics](https://mlsanity.com/paper/2303.09874/) (2023)
- [Generative flow-based warm start of the variational quantum eigensolver](https://mlsanity.com/paper/2507.01726/) (2025)
- [Rotated Mean-Field Variational Inference and Iterative Gaussianization](https://mlsanity.com/paper/2510.07732/) (2025)
- [One-Sided Quantile Coupling for Flow Matching](https://mlsanity.com/paper/2608.00978/) (2026)
- [Kernelised Normalising Flows](https://mlsanity.com/paper/2307.14839/) (2023)
- [MixerFlow: MLP-Mixer meets Normalising Flows](https://mlsanity.com/paper/2310.16777/) (2023)
- [Evaluating Perceptual Distance Models by Fitting Binomial Distributions to Two-Alternative Forced Choice Data](https://mlsanity.com/paper/2403.10390/) (2024)
- [Adaptive Heterogeneous Mixtures of Normalising Flows for Robust Variational Inference](https://mlsanity.com/paper/2510.02056/) (2025)
- [Matricial Free Energy as a Gaussianizing Regularizer: Enhancing Autoencoders for Gaussian Code Generation](https://mlsanity.com/paper/2510.17120/) (2025)
- [Enhancing diffusion models with Gaussianization preprocessing](https://mlsanity.com/paper/2512.21020/) (2025)
- [Stable Global Weighting of Flow Mixtures using Simplex Exponential Moving Average](https://mlsanity.com/paper/2607.03809/) (2026)
- [Linear Independent Component Analysis via Optimal Transport](https://mlsanity.com/paper/2607.14081/) (2026)
- [MUtE: A Dual Framework for Concept Erasure and Counterfactual Interventions](https://mlsanity.com/paper/2609.11253/) (2026)
- [Beyond Random Couplings: Contrastive Noise Alignment in Generative Flows](https://mlsanity.com/paper/2609.18488/) (2026)

## Abstract (arXiv)

Most signal processing problems involve the challenging task of multidimensional probability density function (PDF) estimation. In this work, we propose a solution to this problem by using a family of Rotation-based Iterative Gaussianization (RBIG) transforms. The general framework consists of the sequential application of a univariate marginal Gaussianization transform followed by an orthonormal transform. The proposed procedure looks for differentiable transforms to a known PDF so that the unknown PDF can be estimated at any point of the original domain. In particular, we aim at a zero mean unit covariance Gaussian for convenience. RBIG is formally similar to classical iterative Projection Pursuit (PP) algorithms. However, we show that, unlike in PP methods, the particular class of rotations used has no special qualitative relevance in this context, since looking for interestingness is not a critical issue for PDF estimation. The key difference is that our approach focuses on the univariate part (marginal Gaussianization) of the problem rather than on the multivariate part (rotation). This difference implies that one may select the most convenient rotation suited to each practical application. The differentiability, invertibility and convergence of RBIG are theoretically and experimentally analyzed. Relation to other methods, such as Radial Gaussianization (RG), one-class support vector domain description (SVDD), and deep neural networks (DNN) is also pointed out. The practical performance of RBIG is successfully illustrated in a number of multidimensional problems such as image synthesis, classification, denoising, and multi-information estimation.