# ℓ0 Sparsifying Transform Learning with Efficient Optimal Updates and Convergence Guarantees

arXiv: https://arxiv.org/abs/1501.02859 · HTML: https://mlsanity.com/paper/1501.02859/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Jan 2015.

## Built on by (most cited first, 6 of 6)

- [Model-based Reconstruction with Learning: From Unsupervised to Supervised and Beyond](https://mlsanity.com/paper/2103.14528/) (2021)
- [Differentiating Nonsmooth Solutions to Parametric Monotone Inclusion Problems](https://mlsanity.com/paper/2212.07844/) (2022)
- [Supervised learning of analysis-sparsity priors with automatic differentiation](https://mlsanity.com/paper/2112.07990/) (2021)
- [Multi-layer Clustering-based Residual Sparsifying Transform for Low-dose CT Image Reconstruction](https://mlsanity.com/paper/2203.11565/) (2022)
- [Learning Explicitly Conditioned Sparsifying Transforms](https://mlsanity.com/paper/2403.03168/) (2024)
- [Reconciliation of Statistical and Spatial Sparsity For Robust Image and Image-Set Classification](https://mlsanity.com/paper/2106.00256/) (2021)

## Abstract (arXiv)

Many applications in signal processing benefit from the sparsity of signals in a certain transform domain or dictionary. Synthesis sparsifying dictionaries that are directly adapted to data have been popular in applications such as image denoising, inpainting, and medical image reconstruction. In this work, we focus instead on the sparsifying transform model, and study the learning of well-conditioned square sparsifying transforms. The proposed algorithms alternate between a ℓ0 "norm"-based sparse coding step, and a non-convex transform update step. We derive the exact analytical solution for each of these steps. The proposed solution for the transform update step achieves the global minimum in that step, and also provides speedups over iterative solutions involving conjugate gradients. We establish that our alternating algorithms are globally convergent to the set of local minimizers of the non-convex transform learning problems. In practice, the algorithms are insensitive to initialization. We present results illustrating the promising performance and significant speed-ups of transform learning over synthesis K-SVD in image denoising.