paper-with-me

Papers

On the Theoretical Expressive Power and the Design Space of Higher-Order Graph Transformers

2024-04-04 · Cai Zhou, Rose Yu, Yusu Wang

Graph transformers have recently received significant attention in graph learning, partly due to their ability to capture more global interaction via self-attention. Nevertheless, while higher-order graph neural networks have been reasonably well studied, the exploration of extending graph transformers to higher-order variants is just starting. Both theoretical understanding and empirical results are limited. In this paper, we provide a systematic study of the theoretical expressive power of order-$k$ graph transformers and sparse variants. We first show that, an order-$k$ graph transformer without additional structural information is less expressive than the $k$-Weisfeiler Lehman ($k$-WL) test despite its high computational cost. We then explore strategies to both sparsify and enhance the higher-order graph transformers, aiming to improve both their efficiency and expressiveness. Indeed, sparsification based on neighborhood information can enhance the expressive power, as it provides additional information about input graph structures. In particular, we show that a natural neighborhood-based sparse order-$k$ transformer model is not only computationally efficient, but also expressive -- as expressive as $k$-WL test. We further study several other sparse graph attention models that are computationally efficient and provide their expressiveness analysis. Finally, we provide experimental results to show the effectiveness of the different sparsification strategies.

📄 PDF Abstract BibTeX arXiv:2404.03380

Code (1)

zhouc20/k-transformer 공식 구현 pytorch

Tasks

Graph AttentionGraph Learning

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
LapEigen 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Multi-Head Attention 설명 없음
Adam 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Attention 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

Fine-Grained Expressive Power of Weisfeiler-Leman: A Homomorphism Counting Perspective

2024-10-04 · Junru Zhou, Muhan Zhang

The ability of graph neural networks (GNNs) to count homomorphisms has recently been proposed as a practical and fine-grained measure of their expressive power. Although several existing works have investigated the homom…

Extending the Design Space of Graph Neural Networks by Rethinking Folklore Weisfeiler-Lehman

2023-06-05 · NeurIPS 2023 11 · Jiarui Feng, Lecheng Kong, Hao liu, DaCheng Tao 외

Message passing neural networks (MPNNs) have emerged as the most popular framework of graph neural networks (GNNs) in recent years. However, their expressive power is limited by the 1-dimensional Weisfeiler-Lehman (1-WL)…

Graph Regression

How Powerful are K-hop Message Passing Graph Neural Networks

2022-05-26 · Jiarui Feng, Yixin Chen, Fuhai Li, Anindya Sarkar 외

The most popular design paradigm for Graph Neural Networks (GNNs) is 1-hop message passing -- aggregating information from 1-hop neighbors repeatedly. However, the expressive power of 1-hop message passing is bounded by …

On the Expressive Power of Permutation-Equivariant Weight-Space Networks

2026-02-01 · Adir Dayan, Yam Eitan, Haggai Maron arxiv

Weight-space learning studies neural architectures that operate directly on the parameters of other neural networks. Motivated by the growing availability of pretrained models, recent work has demonstrated the effectiven…

Understanding Truncated Positional Encodings for Graph Neural Networks

2026-06-11 · James Flora, Mitchell Black, Weng-Keen Wong, Amir Nayyeri arxiv

Positional encodings (PEs) enhance the power of graph neural networks (GNNs), both theoretically and empirically. Two of the most popular families of PEs - spectral (e.g., Laplacian eigenspaces, effective resistance) and…