SCRAPL: Scattering Transform with Random Paths for Machine Learning

ICLR 2026 Conference Submission25282 Authors

Keywords: scattering transform, wavelets, stochastic optimization, ddsp, perceptual quality assessment

TL;DR: A stochastic optimization scheme for efficient perceptual quality assessment of deep inverse problems, implemented for differentiable joint time–frequency scattering, with applications to unsupervised sound matching of the Roland TR-808 drum machine.

Abstract:

The Euclidean distance between differentiable wavelet scattering transform coefficients (known as paths) provides informative gradients for perceptual quality assessment of deep inverse problems in computer vision, speech, and audio processing.
However, these transforms are computationally expensive when employed as differentiable loss functions for stochastic gradient descent due to their numerous paths, which significantly limits their use in neural network training.
Against this problem, we propose ``Scattering transform with Random Paths for machine Learning'' (SCRAPL): a stochastic optimization scheme for efficient evaluation of multivariable scattering transforms.
We implement SCRAPL for the joint time–frequency scattering transform (JTFS) which demodulates spectrotemporal patterns at multiple scales and rates, allowing a fine characterization of intermittent auditory textures.
We apply SCRAPL to differentiable digital signal processing (DDSP), specifically, unsupervised sound matching of a granular synthesizer and the Roland TR-808 drum machine.
We also propose an initialization heuristic based on importance sampling, which adapts SCRAPL to the perceptual content of the dataset, improving neural network convergence and evaluation performance.
We make our audio samples available and provide SCRAPL as a Python package.

Primary Area: learning on time series and dynamical systems

Submission Number: 25282