$$ \newcommand{\floor}[1]{\left\lfloor #1 \right\rfloor} \newcommand{\ceil}[1]{\left\lceil #1 \right\rceil} \newcommand{\N}{\mathbb{N}} \newcommand{\R}{\mathbb{R}} \newcommand{\Z}{\mathbb{Z}} \newcommand{\Q}{\mathbb{Q}} \newcommand{\C}{\mathbb{C}} \renewcommand{\L}{\mathcal{L}} \newcommand{\x}{\times} \newcommand{\contra}{\scalebox{1.5}{$\lightning$}} \newcommand{\inner}[2]{\left\langle #1 , #2 \right\rangle} \newcommand{\st}{\text{ such that }} \newcommand{\for}{\text{ for }} \newcommand{\Setcond}[2]{ \left\{\, #1 \mid #2 \, \right\}} \newcommand{\setcond}[2]{\Setcond{#1}{#2}} \newcommand{\seq}[1]{ \left\langle #1 \right\rangle} \newcommand{\Set}[1]{ \left\{ #1 \right\}} \newcommand{\set}[1]{ \Set{#1} } \newcommand{\sgn}{\text{sign}} \newcommand{\halfline}{\vspace{0.5em}} \newcommand{\diag}{\text{diag}} \newcommand{\legn}[2]{\left(\frac{#1}{#2}\right)} \newcommand{\ord}{\text{ord}} \newcommand{\di}{\mathrel{|}} \newcommand{\gen}[1] \newcommand{\irr}{\mathrm{irr }} \renewcommand{\deg}{\mathrm{deg }} \newcommand{\nsgeq}{\trianglelefteq} \newcommand{\nsg}{\triangleleft} \newcommand{\argmin}{\mathrm{argmin}} \newcommand{\argmax}{\mathrm{argmax}} \newcommand{\minimize}{\mathrm{minimize}} \newcommand{\maximize}{\mathrm{maximize}} \newcommand{\subto}{\mathrm{subject\ to}} \newcommand{\DKL}[2]{D_{\mathrm{KL}}\left(#1 \di\di #2\right)} \newcommand{\ReLU}{\mathrm{ReLU}} \newcommand{\E}{\mathsf{E}} \newcommand{\V}{\mathsf{Var}} \newcommand{\Corr}{\mathsf{Corr}} \newcommand{\Cov}{\mathsf{Cov}} \newcommand{\covariance}[1]{\Cov\left(#1\right)} \newcommand{\variance}[1]{\V\left[#1\right]} \newcommand{\variancewith}[1]{\V\left[#1\right]} \newcommand{\expect}[1]{\E\left[#1\right]} \newcommand{\expectwith}[2]{\E_{#1}\left[#2\right]} \renewcommand{\P}{\mathsf{P}} \newcommand{\uniform}[2]{\mathrm{Uniform}\left(#1 \dots #2\right)} \newcommand{\gdist}[2]{\mathcal{N}\left(#1, #2\right)} \DeclarePairedDelimiter{\norm}{\lVert}{\rVert} $$ \everymath{\displaystyle}

An attention Multi-scale co-evolving model for dynamic link prediction | Hyunju Kim

An attention Multi-scale co-evolving model for dynamic link prediction

Guozhen Zhang et al. / WWW / 2023

May 14, 2023

[Abstract]

dynamic link prediction
propose Attentional Multi-scale Co-evolving Network (AMC-Net)
- model multi-scale structural information by a motif-based GNN with multi-scale pooling

[1. Introduction]

Untitled

two types of important information
- temporal information
  - characterizes evolving dynamics of the network
- structural information
  - network topology
⇒ Those two are deeply connected
- microscopic level (node, edge)
- mesoscopic level (groups)
- macroscopic level (whole network)
⇒ temporal dynamics of different structural scales complement one another and are coherent in the meantime
- existing works model the two types of information independently
- to bridge the gaps, this paper presents an AMCNet
The first to study the inherent correlations among the evolving dynamics of different structural scales for dynamic link prediction

[2. Problem formulation]

For a given historical series of snapshots G={G(1), G(2), …, G(T)}, aim to predict the link structure at the future time steps: G={G(T+1), …, G(T+n)}

[3. AMCNET]

[3.1 Multi-scale Representation Learning]

Untitled

motif based pooling
- for a given set of motifs M
- motif-based graphs

[4. Experiments]

Untitled

Untitled

three-node motifs → boosts the AUC
four-node motifs → small performance improvement
three-node motifs are the most informative