graph laplacians for rotation equivariant neural networks
play

GRAPH LAPLACIANS FOR ROTATION EQUIVARIANT NEURAL NETWORKS Master - PowerPoint PPT Presentation

Jun 05, 2023 •176 likes •416 views

GRAPH LAPLACIANS FOR ROTATION EQUIVARIANT NEURAL NETWORKS Master thesis in Computational Science and Engineering School of Basic Sciences Department of Mathematics cole Polytechnique Fdrale de Lausanne Supervised by: Michal

  1. GRAPH LAPLACIANS FOR ROTATION EQUIVARIANT NEURAL NETWORKS Master thesis in Computational Science and Engineering School of Basic Sciences Department of Mathematics École Polytechnique Fédérale de Lausanne Supervised by: Michaël Deferrard (EPFL) Nathanaël Perraudin (SDSC – ETH) Piercesare Secchi (Politecnico di Milano) Pierre Vandergheynst (EPFL) 16.07.19 Martino Milani 1

  2. Rotation Equivariant Filtering 16.07.19 Martino Milani 2

  3. Fourier analysis on the sphere and the spherical harmonics [1] [1] Starry: analytic occultation light curves , Luger R. et al., https://rodluger.github.io/starry/v0.3.0/tutorials/basics1.html 16.07.19 Martino Milani 3

  4. Convolution on the sphere Definition: . [2] Convolutions can be done in the spectral domain as usual: • Complexity of FFT algorithms is . • Convolutions are ro rotation equivariant operations • [2] Computing Fourier Transforms and Convolutions on the 2-Sphere , Driscoll J.R. and Healy D.M., 1994. 16.07.19 Martino Milani 4

  5. Our tool to perform fast spherical convolutions: graphs. Adjacency matrix 16.07.19 Martino Milani 5

  6. Graph convolutions Given a weighted adjacency matrix , define the graph Laplacian to be • The graph Laplacian is symmetric: • Convolving on the graph a signal with a kernel is defined as: • Notice the similarity with 16.07.19 Martino Milani 6

  7. How to build a rotation equivariant graph KEY CONCEPT: if if th then the graph Fourier transform will be rotation equivariant. Heat Kernel Graph (HKG): Belkin et al. proved that with a random sampling scheme, in probability [3] [3] Convergence of Laplacian Eigenmaps , Belkin M. and Nyiogi P,, in Advances in Neural Information Processing Systems 19, 2007. 16.07.19 Martino Milani 7

  8. HEALPix: equiarea sampling scheme The HKG Laplacian eigenvectors well approximate the spherical • harmonics: The dot product of the HKG Laplacian eigenvectors and any sampled • signal well approximates the corresponding Fourier coefficient: 16.07.19 Martino Milani 8

  9. DeepSphere 1.0 DeepSphere [4] is a Spherical Graph Convolutional Neural Network • Each layer implements a po polynomial filter of a sp sparse se HKG Laplacian of a • spherical signal sampled with HEALPix. Filtering is • 0.10 [4] DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications , Perraudin N., Defferrard M., Kacprzak T., Sgier R., 2018 16.07.19 Martino Milani 9

  10. DeepSphere 2.0 DeepSphere 2.0 is an improved version of DeepSphere • Each layer implements a po polynomial filter of a sparse (but no not to too • sparse sp se ) HKG Laplacian of a spherical signal sampled with HEALPix. Filtering is • 0.10 16.07.19 Martino Milani 10

  11. Non-equiarea sampling schemes Figure: left, FEM diffusion of a point source signal. Right, diffusion obtained with the HKG Laplacian 16.07.19 Martino Milani 11

  12. Towards the Finite Element Method (FEM) blem is a discretized version of the weak eigenvalue The Ga Galer erkin pr probl problem, where the ambient space is finite dimensional: 16.07.19 Martino Milani 12

  13. The Finite Element Method (FEM) The Finite Element Method (FEM) is a method to solve the Galerkin problem where the space 𝑊 " is the space of continuous piecewise linear functions defined on a triangulation of the sphere. 16.07.19 Martino Milani 13

  14. The Finite Element Method (FEM) Writing the Galerkin problem times, each time with we obtain the following al aic generalized eigenvalue problem: algebrai 16.07.19 Martino Milani 14

  15. FEM convergence [5] Remark: it’s the on only case (together with the random HKG Laplacian) where we have a convergence theorem. [5] A priori estimates for the FEM approximations to eigenvalues and eigenfunctions of the Laplace-Beltrami operator , Bonito et al., 2017. 16.07.19 Martino Milani 15

  16. FEM Fourier transform Equivalent to finding the decomposition • The eigenvectors are such that: • FEM Fourier transform: • 16.07.19 Martino Milani 16

  17. FEM filtering Filtering a signal with a kernel is defined as the following matrix multiplication, where is the FEM Fourier matrix. 16.07.19 Martino Milani 17

  18. FEM polynomial filtering is not symmetric. • is full. • . • 16.07.19 Martino Milani 18

  19. Lumped FEM Laplacian is not symmetric. • is sparse. • 16.07.19 Martino Milani 19

  20. Lumped FEM Laplacian as a graph Laplacian 16.07.19 Martino Milani 20

  21. Equivariance error and computational time 16.07.19 Martino Milani 21

  22. Conclusions 1) We gained a better understanding of the HKG. 2) We put that knowledge in practice, improving the Equivariance Error of DeepSphere. 3) We investigated different Discrete Laplacians, from Differential Geometry to Numerical Mathematics 4) We used this knowledge to better understand the advantages and limitations of Graph Laplacians when it comes to non uniform sampling of the sphere. 16.07.19 Martino Milani 22

Recommend

Neural Networks with Euclidean Symmetry for Physical Sciences 3D rotation- and

Neural Networks with Euclidean Symmetry for Physical Sciences 3D rotation- and

Neural Networks with Euclidean Symmetry for Physical Sciences 3D rotation- and translation-equivariant convolutional neural networks (for points, meshes, images, ...) Tess Smidt 2018 Alvarez Fellow CSA Summer Series in Computing Sciences

1.09k views • 70 slides
Graph Neural Networks E. Daller, S. Bougleux, and L. Brun April 19, 2018 () Graph Neural

Graph Neural Networks E. Daller, S. Bougleux, and L. Brun April 19, 2018 () Graph Neural

Graph Neural Networks E. Daller, S. Bougleux, and L. Brun April 19, 2018 () Graph Neural Networks April 19, 2018 1 / 10 Different Levels/Steps of Recognition Level 0: Hand made classification (Expert Systems) If x > 0 . 3 and y < 1 . 5

889 views • 10 slides
Learning Discrete Structures for Graph Neural Networks Luca Franceschi , Mathias Niepert,

Learning Discrete Structures for Graph Neural Networks Luca Franceschi , Mathias Niepert,

Learning Discrete Structures for Graph Neural Networks Luca Franceschi , Mathias Niepert, Massimilano Potil, Xiao He Poster later: Pacific Ballroom # 177 Introduction & Motivations Aim: apply Graph Neural Networks (GNN) to settings in which

238 views • 12 slides
Logical Expressiveness of Graph Neural Networks Mikal Monet October 10th, 2019 Millennium

Logical Expressiveness of Graph Neural Networks Mikal Monet October 10th, 2019 Millennium

Logical Expressiveness of Graph Neural Networks Mikal Monet October 10th, 2019 Millennium Institute for Foundational Research on Data, Chile Graph Neural Networks (GNNs) With: Pablo Barcel, Egor Kostylev, Jorge Prez, Juan Reutter,

777 views • 56 slides
Logical Expressiveness of Graph Neural Networks DIG seminar Mikal Monet March 12th, 2020

Logical Expressiveness of Graph Neural Networks DIG seminar Mikal Monet March 12th, 2020

Logical Expressiveness of Graph Neural Networks DIG seminar Mikal Monet March 12th, 2020 Millennium Institute for Foundational Research on Data, Chile Graph Neural Networks (GNNs) With: Pablo Barcel, Egor Kostylev, Jorge Prez, Juan

1.26k views • 86 slides
Graph Neural Networks Xiachong Feng TG 2019-04-08 Relies heavily on A Gentle Introduction

Graph Neural Networks Xiachong Feng TG 2019-04-08 Relies heavily on A Gentle Introduction

Graph Neural Networks Xiachong Feng TG 2019-04-08 Relies heavily on A Gentle Introduction to Graph Neural Networks (Basics, DeepWalk, and GraphSage) Structured deep models: Deep learning on graphs and beyond Representation Learning

4.09k views • 64 slides
Algorithmic investigation of substitution tilings and their associated graph Laplacians Nicole

Algorithmic investigation of substitution tilings and their associated graph Laplacians Nicole

Algorithmic investigation of substitution tilings and their associated graph Laplacians Nicole Harris*, Hayley LeBlanc, Alex Tubbs* Advisor: Professor May Mei Denison University Background Information Aperiodic Tiling An aperiodic tiling

799 views • 37 slides
High Performance Linear System Solvers with Focus on Graph Laplacians Richard Peng Georgia Tech

High Performance Linear System Solvers with Focus on Graph Laplacians Richard Peng Georgia Tech

High Performance Linear System Solvers with Focus on Graph Laplacians Richard Peng Georgia Tech Co-PIs: John Gilbert (UCSB), Gary Miller (CMU) OUTLINE Problem of Lx = b Benchmarks and Evaluations Tree Based Solvers GRAPH

586 views • 21 slides
Graph Neural Network Fang Yuanqiang, 2019/05/18 Graph Neural Network Why GNN? Preliminary

Graph Neural Network Fang Yuanqiang, 2019/05/18 Graph Neural Network Why GNN? Preliminary

Graph Neural Network Fang Yuanqiang, 2019/05/18 Graph Neural Network Why GNN? Preliminary Fixed graph Vanilla Spectral Graph ConvNets ChebyNet GCN CayleyNet Multiple graphs Variable graph

817 views • 46 slides
Outline Evolution of neurocomputing Artificial neural networks Feed forward

Outline Evolution of neurocomputing Artificial neural networks Feed forward

Introduction to Neural Networks and Deep Learning Ling Guan References: S. Haykin, Neural Networks , 2 nd Edition, New Jersey: Prentice Hall, 1. 2004. C. M. Bishop, Neural Networks for Pattern Re cognition, New York: 2. Oxford University

451 views • 17 slides
DAG-GNN: DAG Structure Learning with Graph Neural Networks Yue Yu 1 , Jie Chen 2 , 3 , Tian Gao 3 ,

DAG-GNN: DAG Structure Learning with Graph Neural Networks Yue Yu 1 , Jie Chen 2 , 3 , Tian Gao 3 ,

DAG-GNN: DAG Structure Learning with Graph Neural Networks Yue Yu 1 , Jie Chen 2 , 3 , Tian Gao 3 , Mo Yu 3 1 Department of Mathematics, Lehigh University, USA 2 MIT-IBM Watson AI Lab, USA 3 IBM Research, USA ICML 2019 June 13th, 2019 Background

254 views • 10 slides
Equivariant Transformer Networks (Poster 18) Kai Sheng Tai , Peter Bailis & Gregory Valiant

Equivariant Transformer Networks (Poster 18) Kai Sheng Tai , Peter Bailis & Gregory Valiant

Equivariant Transformer Networks (Poster 18) Kai Sheng Tai , Peter Bailis & Gregory Valiant Stanford University github.com/stanford-futuredata/ equivariant-transformers Goal: Transformation-invariant models How can we learn models that

511 views • 9 slides
Neural Networks 1. Introduction Fall 2017 Neural Networks are taking over! Neural networks

Neural Networks 1. Introduction Fall 2017 Neural Networks are taking over! Neural networks

Neural Networks 1. Introduction Fall 2017 Neural Networks are taking over! Neural networks have become one of the major thrust areas recently in various pattern recognition, prediction, and analysis problems In many problems they have

1.17k views • 91 slides
Neural Networks 0. Logistics Spring 2019 1 Neural Networks are taking over! Neural networks

Neural Networks 0. Logistics Spring 2019 1 Neural Networks are taking over! Neural networks

Neural Networks 0. Logistics Spring 2019 1 Neural Networks are taking over! Neural networks have become one of the major thrust areas recently in various pattern recognition, prediction, and analysis problems In many problems they have

852 views • 33 slides
Graph Neural Networks for Improved El Nio Forecasting Submitted by: Salva Rhling Cachay,

Graph Neural Networks for Improved El Nio Forecasting Submitted by: Salva Rhling Cachay,

Graph Neural Networks for Improved El Nio Forecasting Submitted by: Salva Rhling Cachay, Arthur Fender Coelho Bucker*, Emma Erickson*, Ernest Pokropek*, Willa Potosnak* Mentors: Bjrn Ltjens, Salomey Osei Work motivated by the ProjectX

226 views • 12 slides
Vision and Language Learning with Graph Neural Networks Linchao Zhu 22 Apr, 2020 Recognition,

Vision and Language Learning with Graph Neural Networks Linchao Zhu 22 Apr, 2020 Recognition,

Vision and Language Learning with Graph Neural Networks Linchao Zhu 22 Apr, 2020 Recognition, LEarning, Reasoning UTS CRICOS 00099F Overview RNNs for Image Captioning Transformer for Image Captioning Graph Network for Visual

385 views • 19 slides
Graph Neural Networks for Neutrino Classification Nicholas Choma and Joan Bruna July 18, 2018

Graph Neural Networks for Neutrino Classification Nicholas Choma and Joan Bruna July 18, 2018

Graph Neural Networks for Neutrino Classification Nicholas Choma and Joan Bruna July 18, 2018 Nicholas Choma and Joan Bruna Graph Neural Networks for Neutrino Classification July 18, 2018 1 / 23 Agenda IceCube Experiment 1 Graph Neural

622 views • 29 slides
Learning Semantic Entity Representations with Knowledge Graph and Deep Neural Networks and its

Learning Semantic Entity Representations with Knowledge Graph and Deep Neural Networks and its

Learning Semantic Entity Representations with Knowledge Graph and Deep Neural Networks and its Application to Named Entity Disambiguation Hongzhao Huang 1 and Larry Heck 2 Computer Science Department, Rensselaer Polytechnic Institute 1 Microsoft

243 views • 23 slides
CONVOLUTIONAL AND RECURRENT NEURAL NETWORKS Neural networks Fully connected networks

CONVOLUTIONAL AND RECURRENT NEURAL NETWORKS Neural networks Fully connected networks

CONVOLUTIONAL AND RECURRENT NEURAL NETWORKS Neural networks Fully connected networks Neuron Non-linearity Softmax layer DNN training Loss function and regularization SGD and backprop Learning rate Overfitting

1.8k views • 86 slides
Neural Networks 1. Introduction Spring 2020 1 Neural Networks are taking over! Neural

Neural Networks 1. Introduction Spring 2020 1 Neural Networks are taking over! Neural

Neural Networks 1. Introduction Spring 2020 1 Neural Networks are taking over! Neural networks have become one of the major thrust areas recently in various pattern recognition, prediction, and analysis problems In many problems they

1.63k views • 119 slides
Neural Networks 1. Introduction Spring 2019 1 Neural Networks are taking over! Neural

Neural Networks 1. Introduction Spring 2019 1 Neural Networks are taking over! Neural

Neural Networks 1. Introduction Spring 2019 1 Neural Networks are taking over! Neural networks have become one of the major thrust areas recently in various pattern recognition, prediction, and analysis problems In many problems they

1.57k views • 124 slides
Relaxation and Hopfield Networks Neural Networks Neural Networks - Hopfield 1 Bibliography

Relaxation and Hopfield Networks Neural Networks Neural Networks - Hopfield 1 Bibliography

Relaxation and Hopfield Networks Neural Networks Neural Networks - Hopfield 1 Bibliography Hopfield, J. J., "Neural networks and physical systems with emergent collective computational abilities," Proceedings of the National Academy

367 views • 19 slides
Graph Representation Learning: Where Probability Theory, Data Mining, and Neural Networks Meet

Graph Representation Learning: Where Probability Theory, Data Mining, and Neural Networks Meet

Graph Representation Learning: Where Probability Theory, Data Mining, and Neural Networks Meet Bruno Ribeir eiro Assis istant nt Profess ssor Departm tment ent of Comp mputer er Scienc nce Purdue Unive Pu versit rsity Joint work

612 views • 44 slides
Stochastic Blockmodels meet Graph Neural Networks Nikhil Mehta Lawrence Carin Piyush Rai

Stochastic Blockmodels meet Graph Neural Networks Nikhil Mehta Lawrence Carin Piyush Rai

Stochastic Blockmodels meet Graph Neural Networks Nikhil Mehta Lawrence Carin Piyush Rai I nternational Conference on Machine Learning (ICML) 2019, Long Beach, CA Problem Statement Goal: Learn sparse node embeddings for graphs.

479 views • 9 slides

More recommend