An official website of the United States government
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
iQAN: Fast and Accurate Vector Search with Efficient Intra-Query Parallelism on Multi-Core Architectures
Vector search has drawn a rapid increase of interest in the research community due to its application in novel AI applications. Maximizing its performance is essential for many tasks but remains preliminary understood. In this work, we investigate the root causes of the scalability bottleneck of using intra-query parallelism to speedup the state-of-the-art graph-based vector search systems on multi-core architectures. Our in-depth analysis reveals several scalability challenges from both system and algorithm perspectives. Based on the insights, we propose iQAN, a parallel search algorithm with a set of optimizations that boost convergence, avoid redundant computations, and mitigate synchronization overhead. Our evaluation results on a wide range of real-world datasets show that iQAN achieves up to 37.7× and 76.6× lower latency than state-of-the-art sequential baselines on datasets ranging from a million to a hundred million datasets. We also show that iQAN achieves outstanding scalability as the graph size or the accuracy target increases, allowing it to outperform the state-of-the-art baseline on two billion-scale datasets by up to 16.0× with up to 64 cores.
more »
« less
- PAR ID:
- 10417477
- Publisher / Repository:
- PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming
- Date Published:
- Journal Name:
- PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming
- Page Range / eLocation ID:
- 313 to 328
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Network alignment is a critical steppingstone behind a variety of multi-network mining tasks. Most of the existing methods essentially optimize a Frobenius-like distance or ranking-based loss, ignoring the underlying geometry of graph data. Optimal transport (OT), together with Wasserstein distance, has emerged to be a powerful approach accounting for the underlying geometry explicitly. Promising as it might be, the state-of-the-art OT-based alignment methods suffer from two fundamental limitations, including (1) effectiveness due to the insufficient use of topology and consistency information and (2) scalability due to the non-convex formulation and repeated computationally costly loss calculation. In this paper, we propose a position-aware regularized optimal transport framework for network alignment named PARROT. To tackle the effectiveness issue, the proposed PARROT captures topology information by random walk with restart, with three carefully designed consistency regularization terms. To tackle the scalability issue, the regularized OT problem is decomposed into a series of convex subproblems and can be efficiently solved by the proposed constrained proximal point method with guaranteed convergence. Extensive experiments show that our algorithm achieves significant improvements in both effectiveness and scalability, outperforming the state-of-the-art network alignment methods and speeding up existing OT-based methods by up to 100 times.more » « less
-
Abstract With the aim of analyzing large-sized multidimensional single-cell datasets, we are describing a method for Cosine-based Tanimoto similarity-refined graph for community detection using Leiden’s algorithm (CosTaL). As a graph-based clustering method, CosTaL transforms the cells with high-dimensional features into a weighted k-nearest-neighbor (kNN) graph. The cells are represented by the vertices of the graph, while an edge between two vertices in the graph represents the close relatedness between the two cells. Specifically, CosTaL builds an exact kNN graph using cosine similarity and uses the Tanimoto coefficient as the refining strategy to re-weight the edges in order to improve the effectiveness of clustering. We demonstrate that CosTaL generally achieves equivalent or higher effectiveness scores on seven benchmark cytometry datasets and six single-cell RNA-sequencing datasets using six different evaluation metrics, compared with other state-of-the-art graph-based clustering methods, including PhenoGraph, Scanpy and PARC. As indicated by the combined evaluation metrics, Costal has high efficiency with small datasets and acceptable scalability for large datasets, which is beneficial for large-scale analysis.more » « less
-
Machine learning models exhibit strong performance on datasets with abundant labeled samples. However, for tabular datasets with extremely high d-dimensional features but limited n samples (i.e. d ≫ n), machine learning models struggle to achieve strong performance due to the risk of overfitting. Here, our key insight is that there is often abundant, auxiliary domain information describing input features which can be structured as a heterogeneous knowledge graph (KG). We propose PLATO, a method that achieves strong performance on tabular data with d ≫ n by using an auxiliary KG describing input features to regularize a multilayer perceptron (MLP). In PLATO, each input feature corresponds to a node in the auxiliary KG. In the MLP’s first layer, each input feature also corresponds to a weight vector. PLATO is based on the inductive bias that two input features corresponding to similar nodes in the auxiliary KG should have similar weight vectors in the MLP’s first layer. PLATO captures this inductive bias by inferring the weight vector for each input feature from its corresponding node in the KG via a trainable message-passing function. Across 6 d ≫ n datasets, PLATO outperforms 13 state-of-the-art baselines by up to 10.19%.more » « less
-
We consider in this paper the similarity search problem that retrieves relevant graphs from a graph database under the well-known graph edit distance (GED) constraint. Formally, given a graph database G = {g1, g2, . . . , gn} and a query graph q, we aim to search the graph gi ∈ g such that the graph edit distance between gi and q, GED(gi, q), is within a user-specified GED threshold, τ. In spite of its theoretical significance and wide applicability, the GED-based similarity search problem is challenging in large graph databases due in particular to a large amount of GED computation incurred, which has proven to be NP-hard. In this paper, we propose a parameterized, partition-based GED lower bound that can be instantiated into a series of tight lower bounds towards synergistically pruning false-positive graphs from before costly GED computation is performed. We design an efficient, selectivity-aware algorithm to partition graphs of into highly selective subgraphs. They are further incorporated in a cost-effective, multi-layered indexing structure, ML-Index (Multi-Layered Index), for GED lower bound cross-checking and false-positive graph filtering with theoretical performance guarantees. Experimental studies in real and synthetic graph databases validate the efficiency and effectiveness of ML-Index, which achieves up to an order of magnitude speedup over the state-of-the-art method for similarity search in graph databases.more » « less
