Input Feature Pruning for Accelerating GNN Inference on Heterogeneous Platforms
Input Feature Pruning for Accelerating GNN Inference on Heterogeneous Platforms
复制标题
DOI：
10.1109/hipc56025.2022.00045
复制
发表时间：
2022-12
期刊：
影响因子：
--
通讯作者：
Jason Yik;S. Kuppannagari;Hanqing Zeng;V. Prasanna
中科院分区：
文献类型：
--
作者：
Jason Yik;S. Kuppannagari;Hanqing Zeng;V. Prasanna
图神经网络(GNN)是一种新兴的机器学习模型，它利用结构化的图信息和节点特征，将高维输入数据简化为低维嵌入，从而进行预测。由于聚合邻居信息的复合效应，GNN推断需要来自比预测目标节点多许多倍的原始数据。因此，在异类计算平台上，推理延迟可能在很大程度上受制于甚至在计算开始之前将输入特征数据传输到GPU/加速器的设备间通信成本。在本文中，我们分析了从GNN模型中剪枝输入特征的权衡效果，减少了模型处理的原始数据量，以降低通信延迟为代价，但预期总体模型精度会降低。我们开发贪婪和基于回归的算法来确定保留哪些特征以获得最佳预测精度。我们评估了剪枝模型的变体，发现它们可以减少高达80%的推理延迟，而与非剪枝模型相比，精确度损失不到5%。此外，我们还证明了输入特征剪枝在不同的系统变量(如批次大小和浮点精度)下可以扩展延迟。
Graph Neural Networks (GNNs) are an emerging class of machine learning models which utilize structured graph information and node features to reduce high-dimensional input data to low-dimensional embeddings, from which predictions can be made. Due to the compounding effect of aggregating neighbor information, GNN inferences require raw data from many times more nodes than are targeted for prediction. Thus, on heterogeneous compute platforms, inference latency can be largely subject to the inter-device communication cost of transferring input feature data to the GPU/accelerator before computation has even begun. In this paper, we analyze the trade-off effect of pruning input features from GNN models, reducing the volume of raw data that the model works with to lower communication latency at the expense of an expected decrease in the overall model accuracy. We develop greedy and regression-based algorithms to determine which features to retain for optimal prediction accuracy. We evaluate pruned model variants and find that they can reduce inference latency by up to 80% with an accuracy loss of less than 5% compared to non-pruned models. Furthermore, we show that the latency reductions from input feature pruning can be extended under different system variables such as batch size and floating point precision.