Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks

TL;DR

This paper introduces Causal Anonymous Walks (CAWs) for inductive representation learning in temporal networks, capturing universal network laws without relying on node identities, and demonstrates superior link prediction performance.

Contribution

The paper proposes CAWs and CAW-N, a novel inductive method that automatically captures network motifs and laws, enabling effective learning and prediction in temporal networks.

Findings

CAW-N outperforms previous SOTA methods by 10% AUC on average in inductive link prediction.

CAW-N achieves better results than previous methods in 4 out of 6 networks in transductive setting.

The method supports online training with constant memory and time cost.

Abstract

Temporal networks serve as abstractions of many real-world dynamic systems. These networks typically evolve according to certain laws, such as the law of triadic closure, which is universal in social networks. Inductive representation learning of temporal networks should be able to capture such laws and further be applied to systems that follow the same laws but have not been unseen during the training stage. Previous works in this area depend on either network node identities or rich edge attributes and typically fail to extract these laws. Here, we propose Causal Anonymous Walks (CAWs) to inductively represent a temporal network. CAWs are extracted by temporal random walks and work as automatic retrieval of temporal network motifs to represent network dynamics while avoiding the time-consuming selection and counting of those motifs. CAWs adopt a novel anonymization strategy that replaces node identities with the hitting counts of the nodes based on a set of sampled walks to keep the method inductive, and simultaneously establish the correlation between motifs. We further propose a neural-network model CAW-N to encode CAWs, and pair it with a CAW sampling strategy with constant memory and time cost to support online training and inference. CAW-N is evaluated to predict links over 6 real temporal networks and uniformly outperforms previous SOTA methods by averaged 10% AUC gain in the inductive setting. CAW-N also outperforms previous methods in 4 out of the 6 networks in the transductive setting.