An official website of the United States government
Graph Neural Networks (GNNs) that are based on the message passing (MP) paradigm generally exchange information between 1-hop neighbors to build node representations at each layer. In principle, such networks are not able to capture long-range interactions (LRI) that may be desired or necessary for learning a given task on graphs. Recently, there has been an increasing interest in development of Transformer-based methods for graphs that can consider full node connectivity beyond the original sparse structure, thus enabling the modeling of LRI. However, MP-GNNs that simply rely on 1-hop message passing often fare better in several existing graph benchmarks when combined with positional feature representations, among other innovations, hence limiting the perceived utility and ranking of Transformer-like architectures. Here, we present the Long Range Graph Benchmark (LRGB)1 with 5 graph learning datasets: PascalVOC-SP, COCO-SP, PCQM-Contact, Peptides-func and Peptides-struct that arguably require LRI reasoning to achieve strong performance in a given task. We benchmark both baseline GNNs and Graph Transformer networks to verify that the models which capture long-range dependencies perform significantly better on these tasks. Therefore, these datasets are suitable for benchmarking and exploration of MP-GNNs and Graph Transformer architectures that are intended to capture LRI.
more »
« less
Generalized Relative Neighborhood Graph (GRNG) for Similarity Search
Abstract. Similaritysearchisafundamentalbuildingblockforinformation retrieval on a variety of datasets. The notion of a neighbor is often based on binary considerations, such as the k nearest neighbors. However, considering that data is often organized as a manifold with low intrinsic dimension, the notion of a neighbor must recognize higher-order relationship, to capture neighbors in all directions. Proximity graphs such as the Relative Neighbor Graphs (RNG), use trinary relationships which capture the notion of direc- tion and have been successfully used in a number of applications. However, the current algorithms for computing the RNG, despite widespread use, are approximate and not scalable. This paper proposes a novel type of graph, the Generalized Relative Neighborhood Graph (GRNG) for use in a pivot layer that then guides the efficient and exact construction of the RNG of a set of exemplars. It also shows how to extend this to a multi-layer hier- archy which significantly improves over the state-of-the-art methods which can only construct an approximate RNG.
more »
« less
- Award ID(s):
- 1910530
- PAR ID:
- 10387072
- Date Published:
- Journal Name:
- International Conference on Similarity Search and Applications
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Most state-of-the-art Graph Neural Networks (GNNs) can be defined as a form of graph convolution which can be realized by message passing between direct neighbors or beyond. To scale such GNNs to large graphs, various neighbor-, layer-, or subgraph-sampling techniques are proposed to alleviate the "neighbor explosion" problem by considering only a small subset of messages passed to the nodes in a mini-batch. However, sampling-based methods are difficult to apply to GNNs that utilize many-hops-away or global context each layer, show unstable performance for different tasks and datasets, and do not speed up model inference. We propose a principled and fundamentally different approach, VQ-GNN, a universal framework to scale up any convolution-based GNNs using Vector Quantization (VQ) without compromising the performance. In contrast to sampling-based techniques, our approach can effectively preserve all the messages passed to a mini-batch of nodes by learning and updating a small number of quantized reference vectors of global node representations, using VQ within each GNN layer. Our framework avoids the "neighbor explosion" problem of GNNs using quantized representations combined with a low-rank version of the graph convolution matrix. We show that such a compact low-rank version of the gigantic convolution matrix is sufficient both theoretically and experimentally. In company with VQ, we design a novel approximated message passing algorithm and a nontrivial back-propagation rule for our framework. Experiments on various types of GNN backbones demonstrate the scalability and competitive performance of our framework on large-graph node classification and link prediction benchmarks.more » « less
-
Abstract What are the unavoidable induced subgraphs of graphs with large treewidth? It is well‐known that the answer must include a complete graph, a complete bipartite graph, all subdivisions of a wall and line graphs of all subdivisions of a wall (we refer to these graphs as the “basic treewidth obstructions”). So it is natural to ask whether graphs excluding the basic treewidth obstructions as induced subgraphs have bounded treewidth. Sintiari and Trotignon answered this question in the negative. Their counterexamples, the so‐called “layered wheels,” contain wheels, where awheelconsists of ahole(i.e., an induced cycle of length at least four) along with a vertex with at least three neighbors in the hole. This leads one to ask whether graphs excluding wheels and the basic treewidth obstructions as induced subgraphs have bounded treewidth. This also turns out to be false due to Davies' recent example of graphs with large treewidth, no wheels and no basic treewidth obstructions as induced subgraphs. However, in Davies' example there exist holes and vertices (outside of the hole) with two neighbors in them. Here we prove that a hole with a vertex with at least two neighbors in it is inevitable in graphs with large treewidth and no basic obstruction. Our main result is that graphs in which every vertex has at most one neighbor in every hole (that does not contain it) and with the basic treewidth obstructions excluded as induced subgraphs have bounded treewidth.more » « less
-
Algorithms often carry out equally many computations for “easy” and “hard” problem instances. In particular, algorithms for finding nearest neighbors typically have the same running time regardless of the particular problem instance. In this paper, we consider the approximate k-nearest-neighbor problem, which is the problem of finding a subset of O(k) points in a given set of points that contains the set of k nearest neighbors of a given query point. We pro- pose an algorithm based on adaptively estimating the distances, and show that it is essentially optimal out of algorithms that are only allowed to adaptively estimate distances. We then demonstrate both theoretically and experimentally that the algorithm can achieve significant speedups relative to the naive method.more » « less
-
Effective vector representation models, e.g., word2vec and node2vec, embed real-world objects such as images and documents in high dimensional vector space. In the meanwhile, the objects are often associated with attributes such as timestamps and prices. Many scenarios need to jointly query the vector representations of the objects together with their attributes. These queries can be formalized as range-filtering approximate nearest neighbor search (ANNS) queries. Specifically, given a collection of data vectors, each associated with an attribute value whose domain has a total order. The range-filtering ANNS consists of a query range and a query vector. It finds the approximate nearest neighbors of the query vector among all the data vectors whose attribute values fall in the query range. Existing approaches suffer from a rapidly degrading query performance when the query range width shifts. The query performance can be optimized by a solution that builds an ANNS index for every possible query range; however, the index time and index size become prohibitive -- the number of query ranges is quadratic to the number n of data vectors. To overcome these challenges, for the query range contains all attribute values smaller than a user-provided threshold, we design a structure called the segment graph whose index time and size are the same as a single ANNS index, yet can losslessly compress the n ANNS indexes, reducing the indexing cost by a factor of Ω(n). To handle general range queries, we propose a 2D segment graph with average-case index size O(n log n) to compress n segment graphs, breaking the quadratic barrier. Extensive experiments conducted on real-world datasets show that our proposed structures outperformed existing methods significantly; our index also exhibits superior scalability.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1007/978-3-031-17849-8_11
