FiGURe: Simple and Efficient Unsupervised Node Representations with Filter Augmentations
This work addresses the problem of inefficient and limited unsupervised node representation learning for graph-based tasks, offering computational improvements and better performance on both homophilic and heterophilic datasets.
The paper tackles the limitation of existing unsupervised node representation methods that rely on low-pass filter augmentations, which perform poorly on tasks requiring different parts of the eigen-spectrum, by proposing a simple filter-based augmentation method that captures various eigen-spectrum parts and achieves an average gain of up to 4.4% compared to state-of-the-art models across datasets.
Unsupervised node representations learnt using contrastive learning-based methods have shown good performance on downstream tasks. However, these methods rely on augmentations that mimic low-pass filters, limiting their performance on tasks requiring different eigen-spectrum parts. This paper presents a simple filter-based augmentation method to capture different parts of the eigen-spectrum. We show significant improvements using these augmentations. Further, we show that sharing the same weights across these different filter augmentations is possible, reducing the computational load. In addition, previous works have shown that good performance on downstream tasks requires high dimensional representations. Working with high dimensions increases the computations, especially when multiple augmentations are involved. We mitigate this problem and recover good performance through lower dimensional embeddings using simple random Fourier feature projections. Our method, FiGURe achieves an average gain of up to 4.4%, compared to the state-of-the-art unsupervised models, across all datasets in consideration, both homophilic and heterophilic. Our code can be found at: https://github.com/microsoft/figure.