Scattering Statistics of Generalized Spatial Poisson Point Processes

@inproceedings{Perlmutter2022ScatteringSO,
  title={Scattering Statistics of Generalized Spatial Poisson Point Processes},
  author={Michael Perlmutter and Jieqian He and Matthew J. Hirn},
  booktitle={ICASSP},
  year={2022}
}
We present a machine learning model for the analysis of randomly generated discrete signals, which we model as the points of a homogeneous or inhomogeneous, compound Poisson point process. Like the wavelet scattering transform introduced by S. Mallat, our construction is a mathematical model of convolutional neural networks and is naturally invariant to translations and reflections. Our model replaces wavelets with Gabor-type measurements and therefore decouples the roles of scale and frequency… 

Figures from this paper

References

SHOWING 1-10 OF 35 REFERENCES
Intermittent process analysis with scattering moments
Scattering moments provide nonparametric models of random processes with stationary increments. They are expected values of random variables computed with a nonexpansive operator, obtained by
Statistical learning of geometric characteristics of wireless networks
TLDR
The scattering moments can capture similar geometry information as the baseline approach, and can reach even better performance, especially for non-local marking functions, as well as for stochastic geometry and analysis of wireless networks.
Solid Harmonic Wavelet Scattering for Predictions of Molecule Properties
TLDR
A machine learning algorithm using Gaussian-type orbital functions to create surrogate electronic densities of the molecule from which invariant "solid harmonic scattering coefficients" that account for different types of interactions at different scales are computed.
Invariant Scattering Convolution Networks.
  • Joan Bruna, S. Mallat
  • Computer Science
    IEEE transactions on pattern analysis and machine intelligence
  • 2012
TLDR
The mathematical analysis of wavelet scattering networks explain important properties of deep convolution networks for classification.
Multiscale sparse microcanonical models
TLDR
This paper introduces microcanonical measures whose energy vector is given by a wavelet scattering transform, built by cascading wavelet decompositions and point-wise nonlinearities, and demonstrates the efficiency of sparse multiscale micro canonical measures on several processes and real data exhibiting long-range interactions.
Combined scattering for rotation invariant texture analysis
TLDR
A combined scattering representation for texture classification is introduced, which is invariant to rotations and stable to deformations, and compared with state-of-the-art algorithms, with a nearest neighbor classifier.
Classification with scattering operators
A scattering vector is a local descriptor including multiscale and multi-direction co-occurrence information. It is computed with a cascade of wavelet decompositions and complex modulus. This
Deep Scattering Spectrum
A scattering transform defines a locally translation invariant representation which is stable to time-warping deformation. It extends MFCC representations by computing modulation spectrum
Solid Harmonic Wavelet Scattering: Predicting Quantum Molecular Energy from Invariant Descriptors of 3D Electronic Densities
TLDR
An application of solid harmonic scattering invariants to the estimation of quantum molecular energies, which are also invariant to rigid motion and stable with respect to deformations, is studied.
Rigid-Motion Scattering for Texture Classification
TLDR
A rigid-motion scattering computes adaptive invariants along translations and rotations, with a deep convolutional network, that preserves joint rotation and translation information, while providing global invariants at any desired scale.
...
...