Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Data-Adaptive Graph Framelets with Generalized Vanishing Moments for Graph Signal Processing (2309.03537v2)

Published 7 Sep 2023 in eess.SP, cs.LG, and math.FA

Abstract: In this paper, we propose a novel and general framework to construct tight framelet systems on graphs with localized supports based on hierarchical partitions. Our construction provides parametrized graph framelet systems with great generality based on partition trees, by which we are able to find the size of a low-dimensional subspace that best fits the low-rank structure of a family of signals. The orthogonal decomposition of subspaces provides a key ingredient for the definition of "generalized vanishing moments" for graph framelets. In a data-adaptive setting, the graph framelet systems can be learned by solving an optimization problem on Stiefel manifolds with respect to our parameterization. Moreover, such graph framelet systems can be further improved by solving a subsequent optimization problem on Stiefel manifolds, aiming at providing the utmost sparsity for a given family of graph signals. Experimental results show that our learned graph framelet systems perform superiorly in non-linear approximation and denoising tasks.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (45)
  1. D. I. Shuman, S. K. Narang, P. Frossard, A. Ortega, and P. Vandergheynst, “The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains,” IEEE Signal Processing Magazine, vol. 30, no. 3, pp. 83–98, 2013.
  2. A. Ortega, P. Frossard, J. Kovačević, J. M. Moura, and P. Vandergheynst, “Graph signal processing: Overview, challenges, and applications,” Proceedings of the IEEE, vol. 106, no. 5, pp. 808–828, 2018.
  3. S. K. Narang and A. Ortega, “Perfect reconstruction two-channel wavelet filter banks for graph structured data,” IEEE Transactions on Signal Processing, vol. 60, no. 6, pp. 2786–2799, 2012.
  4. ——, “Compact support biorthogonal wavelet filterbanks for arbitrary undirected graphs,” IEEE Transactions on Signal Processing, vol. 61, no. 19, pp. 4673–4685, 2013.
  5. A. Sakiyama and Y. Tanaka, “Oversampled graph laplacian matrix for graph filter banks,” IEEE Transactions on Signal Processing, vol. 62, no. 24, pp. 6425–6437, 2014.
  6. H. Q. Nguyen and M. N. Do, “Downsampling of signals on graphs via maximum spanning trees,” IEEE Transactions on Signal Processing, vol. 63, no. 1, pp. 182–191, 2014.
  7. X. Zheng, Y. Y. Tang, and J. Zhou, “A framework of adaptive multiscale wavelet decomposition for signals on undirected graphs,” IEEE Transactions on Signal Processing, vol. 67, no. 7, pp. 1696–1711, 2019.
  8. J. You and L. Yang, “Perfect reconstruction two-channel filter banks on arbitrary graphs,” Applied and Computational Harmonic Analysis, vol. 65, pp. 296–321, 2023.
  9. E. Pavez, B. Girault, A. Ortega, and P. A. Chou, “Two channel filter banks on arbitrary graphs with positive semi definite variation operators,” IEEE Transactions on Signal Processing, vol. 71, pp. 917–932, 2023.
  10. S. Li, Y. Jin, and D. I. Shuman, “Scalable m𝑚mitalic_m-channel critically sampled filter banks for graph signals,” IEEE Transactions on Signal Processing, vol. 67, no. 15, pp. 3954–3969, 2019.
  11. N. Tremblay and P. Borgnat, “Subgraph-based filterbanks for graph signals,” IEEE Transactions on Signal Processing, vol. 64, no. 15, pp. 3827–3840, 2016.
  12. A. Sakiyama, K. Watanabe, Y. Tanaka, and A. Ortega, “Two-channel critically sampled graph filter banks with spectral domain sampling,” IEEE Transactions on Signal Processing, vol. 67, no. 6, pp. 1447–1460, 2019.
  13. M. S. Kotzagiannidis and P. L. Dragotti, “Splines and wavelets on circulant graphs,” Applied and Computational Harmonic Analysis, vol. 47, no. 2, pp. 481–515, 2019.
  14. A. Miraki and H. Saeedi-Sourck, “Spline graph filter bank with spectral sampling,” Circuits, Systems, and Signal Processing, vol. 40, no. 11, pp. 5744–5758, 2021.
  15. ——, “A modified spline graph filter bank,” Circuits, Systems, and Signal Processing, vol. 40, pp. 2025–2035, 2021.
  16. V. N. Ekambaram, G. C. Fanti, B. Ayazifar, and K. Ramchandran, “Spline-like wavelet filterbanks for multiresolution analysis of graph-structured data,” IEEE Transactions on Signal and Information Processing over Networks, vol. 1, no. 4, pp. 268–278, 2015.
  17. J. You and L. Yang, “Spline-like wavelet filterbanks with perfect reconstruction on arbitrary graphs,” IEEE Transactions on Signal and Information Processing over Networks, 2023.
  18. D. K. Hammond, P. Vandergheynst, and R. Gribonval, “Wavelets on graphs via spectral graph theory,” Applied and Computational Harmonic Analysis, vol. 30, no. 2, pp. 129–150, 2011.
  19. B. Dong, “Sparse representation on graphs by tight wavelet frames and applications,” Applied and Computational Harmonic Analysis, vol. 42, no. 3, pp. 452–479, 2017.
  20. H. Behjat, U. Richter, D. Van De Ville, and L. Sörnmo, “Signal-adapted tight frames on graphs,” IEEE Transactions on Signal Processing, vol. 64, no. 22, pp. 6017–6029, 2016.
  21. N. Leonardi and D. Van De Ville, “Tight wavelet frames on multislice graphs,” IEEE Transactions on Signal Processing, vol. 61, no. 13, pp. 3357–3367, 2013.
  22. D. B. Tay, Y. Tanaka, and A. Sakiyama, “Almost tight spectral graph wavelets with polynomial filters,” IEEE Journal of Selected Topics in Signal Processing, vol. 11, no. 6, pp. 812–824, 2017.
  23. M. Crovella and E. Kolaczyk, “Graph wavelets for spatial traffic analysis,” in IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No. 03CH37428), vol. 3.   IEEE, 2003, pp. 1848–1857.
  24. M. Gavish, B. Nadler, and R. R. Coifman, “Multiscale wavelets on trees, graphs and high dimensional data: Theory and applications to semi supervised learning,” in ICML, 2010.
  25. C. K. Chui, F. Filbir, and H. N. Mhaskar, “Representation of functions on big data: graphs and trees,” Applied and Computational Harmonic Analysis, vol. 38, no. 3, pp. 489–509, 2015.
  26. C. K. Chui, H. Mhaskar, and X. Zhuang, “Representation of functions on big data associated with directed graphs,” Applied and Computational Harmonic Analysis, vol. 44, no. 1, pp. 165–188, 2018.
  27. J. Li, R. Zheng, H. Feng, M. Li, and X. Zhuang, “Permutation Equivariant Graph Framelets for Heterophilous Graph Learning,” arXiv e-prints, p. arXiv:2306.04265, Jun. 2023.
  28. B. Han, “Framelets and wavelets,” Algorithms, Analysis, and Applications, Applied and Numerical Harmonic Analysis. Birkhäuser xxxiii Cham, 2017.
  29. E. J. Candès and D. L. Donoho, “New tight frames of curvelets and optimal representations of objects with piecewise c2superscript𝑐2c^{2}italic_c start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT singularities,” Communications on Pure and Applied Mathematics: A Journal Issued by the Courant Institute of Mathematical Sciences, vol. 57, no. 2, pp. 219–266, 2004.
  30. G. Kutyniok and D. Labate, “Introduction to shearlets,” Shearlets: Multiscale Analysis for Multivariate Data, pp. 1–38, 2012.
  31. B. Han and X. Zhuang, “Smooth affine shear tight frames with mra structure,” Applied and Computational Harmonic Analysis, vol. 39, no. 2, pp. 300–338, 2015.
  32. B. Han, Z. Zhao, and X. Zhuang, “Directional tensor product complex tight framelets with low redundancy,” Applied and Computational Harmonic Analysis, vol. 41, no. 2, pp. 603–637, 2016.
  33. X. Zhuang, “Digital affine shear transforms: fast realization and applications in image/video processing,” SIAM Journal on Imaging Sciences, vol. 9, no. 3, pp. 1437–1466, 2016.
  34. Z. Che and X. Zhuang, “Digital affine shear filter banks with 2-layer structure and their applications in image processing,” IEEE Transactions on Image Processing, vol. 27, no. 8, pp. 3931–3941, 2018.
  35. B. Han, T. Li, and X. Zhuang, “Directional compactly supported box spline tight framelets with simple geometric structure,” Applied Mathematics Letters, vol. 91, pp. 213–219, 2019.
  36. B. Han, Q. Mo, Z. Zhao, and X. Zhuang, “Directional compactly supported tensor product complex tight framelets with applications to image denoising and inpainting,” SIAM Journal on Imaging Sciences, vol. 12, no. 4, pp. 1739–1771, 2019.
  37. N. Sharon and Y. Shkolnisky, “A class of laplacian multiwavelets bases for high-dimensional data,” Applied and Computational Harmonic Analysis, vol. 38, no. 3, pp. 420–451, 2015.
  38. J. Li, H. Feng, and X. Zhuang, “Convolutional neural networks for spherical signal processing via area-regular spherical haar tight framelets,” IEEE Transactions on Neural Networks and Learning Systems, 2022.
  39. S. G. Mallat, “A theory for multiresolution signal decomposition: the wavelet representation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, no. 7, pp. 674–693, 1989.
  40. A. Ron and Z. Shen, “Affine systems in L2⁢(Rd)subscript𝐿2superscript𝑅𝑑{L_{2}(R^{d})}italic_L start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT ( italic_R start_POSTSUPERSCRIPT italic_d end_POSTSUPERSCRIPT ): the analysis of the analysis operator,” Journal of Functional Analysis, vol. 148, no. 2, pp. 408–447, 1997.
  41. X. Zheng, B. Zhou, Y. G. Wang, and X. Zhuang, “Decimated framelet system on graphs and fast g-framelet transforms,” The Journal of Machine Learning Research, vol. 23, no. 1, pp. 787–854, 2022.
  42. Y. G. Wang and X. Zhuang, “Tight framelets and fast framelet filter bank transforms on manifolds,” Applied and Computational Harmonic Analysis, vol. 48, no. 1, pp. 64–95, 2020.
  43. N. T. Trendafilov, “On the l1subscript𝑙1l_{1}italic_l start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT procrustes problem,” Future Generation Computer Systems, vol. 19, no. 7, pp. 1177–1186, 2003.
  44. J. Townsend, N. Koep, and S. Weichwald, “Pymanopt: A python toolbox for optimization on manifolds using automatic differentiation,” Journal of Machine Learning Research, vol. 17, no. 137, pp. 1–5, 2016.
  45. N. Perraudin, J. Paratte, D. Shuman, L. Martin, V. Kalofolias, P. Vandergheynst, and D. K. Hammond, “Gspbox: A toolbox for signal processing on graphs,” arXiv preprint arXiv:1408.5781, 2014.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now