Merge pull request #475 from grlee77/paper_rev2

JOSS: mention the related package, Kymatio, as well.

Author: rgommers

doc/paper/paper.bib

@@ -85,6 +85,36 @@ @Article{scikit-image
  doi = {10.7717/peerj.453}
 }
 
+@article{kymatio,
+  author    = {Mathieu Andreux and
+               Tom{\'{a}}s Angles and
+               Georgios Exarchakis and
+               Roberto Leonarduzzi and
+               Gaspar Rochette and
+               Louis Thiry and
+               John Zarka and
+               St{\'{e}}phane Mallat and
+               Joakim And{\'{e}}n and
+               Eugene Belilovsky and
+               Joan Bruna and
+               Vincent Lostanlen and
+               Matthew J. Hirn and
+               Edouard Oyallon and
+               Sixhin Zhang and
+               Carmine{-}Emanuele Cella and
+               Michael Eickenberg},
+  title     = {Kymatio: Scattering Transforms in Python},
+  journal   = {CoRR},
+  volume    = {abs/1812.11214},
+  year      = {2018},
+  url       = {http://arxiv.org/abs/1812.11214},
+  archivePrefix = {arXiv},
+  eprint    = {1812.11214},
+  timestamp = {Wed, 02 Jan 2019 14:40:18 +0100},
+  biburl    = {https://dblp.org/rec/bib/journals/corr/abs-1812-11214},
+  bibsource = {dblp computer science bibliography, https://dblp.org}
+}
+
 @Misc{odl,
   author       = {Jonas Adler and
                   Holger Kohr and

doc/paper/paper.md

@@ -76,11 +76,17 @@ or optional dependency by a number of other software projects. For example,
 it has enabled wavelet-based image denoising in scikit-image [@scikit-image].
 The Operator Discretization Library (ODL) [@odl] uses PyWavelets to enable
 wavelet-based regularization in iterative inverse problems such as computed
-tomography image reconstruction. A number of common 1D demo signals used in
-the literature and in the manuscript by Stephan Mallat [@mallat2008wavelet] are
-provided for use in teaching and for purposes of reproducible research. The
-source code for ``PyWavelets`` has been archived to Zenodo with the linked DOI:
-[@pywavelets_zenodo].
+tomography image reconstruction. Another related package which is independent
+of ``PyWavelets`` is Kymatio, which implements the wavelet scattering
+transform in 1D-3D [@kymatio]. The current implementation in Kymatio uses
+non-separable 2D and 3D wavelets defined in the frequency domain and is well
+suited to signal classification tasks, but does not have a simple inverse
+transform like the standard discrete wavelet transform.
+
+A number of common 1D demo signals used in the literature and in the manuscript
+by Stephan Mallat [@mallat2008wavelet] are provided for use in teaching and for
+purposes of reproducible research. The source code for ``PyWavelets`` has been
+archived to Zenodo with the linked DOI: [@pywavelets_zenodo].
 
 # Acknowledgements