arXiv Open Access 2021

Towards Robust Graph Contrastive Learning

Nikola Jovanović Zhao Meng Lukas Faber Roger Wattenhofer
Lihat Sumber

Abstrak

We study the problem of adversarially robust self-supervised learning on graphs. In the contrastive learning framework, we introduce a new method that increases the adversarial robustness of the learned representations through i) adversarial transformations and ii) transformations that not only remove but also insert edges. We evaluate the learned representations in a preliminary set of experiments, obtaining promising results. We believe this work takes an important step towards incorporating robustness as a viable auxiliary task in graph contrastive learning.

Penulis (4)

N

Nikola Jovanović

Z

Zhao Meng

L

Lukas Faber

R

Roger Wattenhofer

Format Sitasi

Jovanović, N., Meng, Z., Faber, L., Wattenhofer, R. (2021). Towards Robust Graph Contrastive Learning. https://arxiv.org/abs/2102.13085

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2021
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓