Timezone: »

 
Poster
Bandit Samplers for Training Graph Neural Networks
Ziqi Liu · Zhengwei Wu · Zhiqiang Zhang · Jun Zhou · Shuang Yang · Le Song · Yuan Qi

Wed Dec 09 09:00 AM -- 11:00 AM (PST) @ Poster Session 3 #756
in Poster Session 4 »

Several sampling algorithms with variance reduction have been proposed for accelerating the training of Graph Convolution Networks (GCNs). However, due to the intractable computation of optimal sampling distribution, these sampling algorithms are suboptimal for GCNs and are not applicable to more general graph neural networks (GNNs) where the message aggregator contains learned weights rather than fixed weights, such as Graph Attention Networks (GAT). The fundamental reason is that the embeddings of the neighbors or learned weights involved in the optimal sampling distribution are \emph{changing} during the training and \emph{not known a priori}, but only \emph{partially observed} when sampled, thus making the derivation of an optimal variance reduced samplers non-trivial. In this paper, we formulate the optimization of the sampling variance as an adversary bandit problem, where the rewards are related to the node embeddings and learned weights, and can vary constantly. Thus a good sampler needs to acquire variance information about more neighbors (exploration) while at the same time optimizing the immediate sampling variance (exploit). We theoretically show that our algorithm asymptotically approaches the optimal variance within a factor of 3. We show the efficiency and effectiveness of our approach on multiple datasets.

Author Information

Ziqi Liu (Ant Group)
Zhengwei Wu (Ant Financial)
Zhiqiang Zhang (Ant Financial Services Group)
Jun Zhou (Ant Financial)
Shuang Yang (Ant Financial)
Le Song (Ant Financial Services Group)
Yuan Qi (Ant Financial Services Group)

More from the Same Authors