arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.18042cs.LG
收藏

用于图摘要的终身图学习

Lifelong Graph Learning for Graph Summarization

Ulm University(乌尔姆大学);University of Essex(埃塞克斯大学)

Jonatan Frank, Marcel Hoffmann, Nicolas Lell, David Richerby, Ansgar Scherp

AI总结:

研究神经网络在终身图摘要中的应用,通过Graph-MLP等模型在时序Web图上实验,发现网络主要依赖1-hop信息且跨十年应用时因异构性增加导致准确率显著下降。

摘要:

由于所建模信息的异构性及其随时间的变化,对Web图进行摘要是一项具有挑战性的任务。我们研究了神经网络在终身图摘要中的应用。假设我们在某一时刻观察到Web图,我们训练网络对图顶点进行摘要。我们将这个训练好的网络应用于在下一个时间点对已变化图的顶点进行摘要。随后,我们继续训练和评估网络以执行终身图摘要。我们使用GNNs Graph-MLP和GraphSAINT,以及一个MLP基线,来对时序图进行摘要。我们比较了$1$-hop和$2$-hop摘要。我们通过测量神经网络的向后和向前迁移以及遗忘率,研究了重用先前快照参数的影响。我们在2012年和2022年采样的、具有超过$100$M条边的Web图的十个每周快照上进行了广泛实验,结果表明,所有网络主要使用$1$-hop信息来确定摘要,即使在执行$2$-hop摘要时也是如此。由于Web图的异构性,在某些快照中,$2$-hop摘要产生的顶点摘要数量是$1$-hop摘要的十倍以上。当使用在2012年最后一个快照上训练的网络并将其应用于2022年的第一个快照时,我们观察到准确率急剧下降。我们将这种跨越十年时间跨度的下降归因于2022年Web图异构性的急剧增加。

英文摘要:

Summarizing web graphs is challenging due to the heterogeneity of the modeled information and its changes over time. We investigate the use of neural networks for lifelong graph summarization. Assuming we observe the web graph at a certain time, we train the networks to summarize graph vertices. We apply this trained network to summarize the vertices of the changed graph at the next point in time. Subsequently, we continue training and evaluating the network to perform lifelong graph summarization. We use the GNNs Graph-MLP and GraphSAINT, as well as an MLP baseline, to summarize the temporal graphs. We compare $1$-hop and $2$-hop summaries. We investigate the impact of reusing parameters from a previous snapshot by measuring the backward and forward transfer and the forgetting rate of the neural networks. Our extensive experiments on ten weekly snapshots of a web graph with over $100$M edges, sampled in 2012 and 2022, show that all networks predominantly use $1$-hop information to determine the summary, even when performing $2$-hop summarization. Due to the heterogeneity of web graphs, in some snapshots, the $2$-hop summary produces over ten times more vertex summaries than the $1$-hop summary. When using the network trained on the last snapshot from 2012 and applying it to the first snapshot of 2022, we observe a strong drop in accuracy. We attribute this drop over the ten-year time warp to the strongly increased heterogeneity of the web graph in 2022.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑