Graph Signals and Graph Signal Processing

Data

Some extra notes of GSO : here
graph convolutions

File status type Lectures
we can use GNNs to solve feature-aware semi-supervised learning problems complete 💡
we can represent any analytic function with convolutional graph filters complete 🧮
unweighted graph complete 💡
unsupervised empty 💡
undirected graph complete 💡
total variation energy complete 💡
the spectral representation of a graph filter is independent from the graph complete 💡
the spectral graph filter operates on a signal pointwise complete 🧮
symmetric laplacian complete 💡
supervised learning empty 💡
stochastic block model complete 💡
statistical risk minimization problem complete 💡
stable graph filter complete 💡
stability-discriminability tradeoff for Lipschitz filters complete 💡
stability and size tradeoff for realistic sparsity pattern considerations setting in progress 💡
spectral representation of a convolutional graph filter complete 💡
spectral graph filter complete 💡
spectral embedding complete 💡
spectral clustering empty 💡
sometimes spectral algorithms fail complete 💡
signal to noise ratio complete 💡
relative perturbations complete 💡
relative perturbation edge changes are tied to node degree complete 💡
readout layer complete 💡
random walk matrix complete 💡
random walk laplacian complete 💡
random graphs in a gin are good for graph isomorphism complete 💡
quasi-symmetry complete 💡
operator distance modulo permutations complete
operator dilation complete 💡
normalized graph laplacian complete 💡
normalized adjacency matrix complete 💡
node-level task complete 💡
neighborhood (graph) complete 💡
network diffusion process complete 💡
multi-layer graph perceptron complete 💡
message passing neural network complete 💡
linear graph filter complete 💡
leaky ReLU complete 💡
inverse graph fourier transform complete 💡
interpretation of the symmetric graph laplacian complete 💡
interpretation of the graph laplacian complete 💡
    integral lipschitz filters are stable to dilations complete 🧮
    integral Lipschitz filter complete 💡
    inductive learning complete 💡
    information theoretic threshold complete 💡
    hypothesis class complete 💡
    homomorphism density complete 💡
    graph complete 💡
    graph-level problem complete 💡
    graph signals complete 💡
    graph signal processing problem complete 💡
    graph shift operator complete 💡
    graph perceptron complete 💡
    graph isomorphism complete 💡
    graph laplacian empty 💡
    graph isomorphism is not solvable in polynomial time complete 💡
      graph homomorphism complete 💡
      graph fourier transform complete 💡
      graph convolutions are stable to perturbations in the data and coefficients complete 💡
      graph convolutional network complete 💡
      graph convolution complete 💡
      graph automorphism complete 💡
      graph attention model complete 💡
      graph SAGE complete 💡
      fully connected readout layer complete 💡
      filter permutation invariance complete 💡
      feature-aware spectral embeddings complete 💡
      eigenvector misalignment complete 💡
      discriminability of a graph filter in progress 💡
      degree matrix complete 💡
      cycle homomorphism density is given by the trace of the adjacency matrix complete 💡
      coordinate representation complete 💡
      convolutional graph filters are shift equivariant complete 🧮
      convolutional graph filters are permutation equivariant complete 🧮
      convolutional graph filters are local complete 🧮
      convolutional filter bank complete 💡
      contextual stochastic block model empty 💡
      conditions for finding a convolutional graph filter complete 🧮
      computational graph complete 💡
      compressed sparse row representation complete 💡
      color refinement algorithm in progress 💡
      chebyshev polynomials are orthogonal complete 💡
      chebyshev equioscillation theorem complete 🧮
      balanced stochastic block model complete 💡
      approximation of heaviside functions using convolutional graph filters empty 🧮
      analytic function complete 💡
      almost exact recovery complete 💡
      almost exact recovery is impossible when the signal to noise ratio is less than the threshold complete 🧮
      aggregation readout layer complete 💡
      adjacency matrix complete 💡
      Weisfeiler-Leman Graph Isomorphism Test in progress 💡
      We can verify whether graphs without node features and different laplacian eigenvalues are not isomorphic complete 🧮
      MPNNs can be expressed as graph convolutions complete 💡
      Lipschitz filters are stable to additive perturbations complete 🧮
      Lipschitz continuous complete 💡
      Graph Neural Networks complete 💡
      Graph Isomorphism Network complete 💡
      GNNs perform better than their constituent filters complete 💡
      GNNs inherit stability from their layers complete 🧮
      GINs are maximally powerful for anonymous input graphs complete 💡
      GCN layers can be written as graph convolutions complete 💡
      Empirical risk minimization problem empty 💡