We thank the reviewers for their time and valuable feedback. "scalability is particularly appealing", "theoretical analysis is great, substantial, valid, correct", experiments are We believe these clarifications, together with our new analyses, resolve all key issues raised. As suggested by reviewers, we will carefully discuss this in the final version. We are not claiming novelty in "the idea of using local subgraphs to compute node GNN on an entire graph together with MAML, performs 42.5% worse than G-M's variant is that it uses the's performance can vary with local subgraph size We will include this analysis in our final version. Empirically, we find the subgraph construction takes 14.7% of training time, and this can be's comment of "evaluating each node label individually", we note that each mini-batch consists We will include the full study in the final version. Disjoint Labels" problem, each task defines an N -size label set, and samples K nodes for each label N

This paper proposes a meta-learning method for graph data. The use of the local subgraphs provides more flexibility and allows us to adopt the same framework to different scenarios. The effectiveness of the proposed method is supported by theory and experiments. However, Meta-GNN also computes the nodes representations using a sub-graph based on the number of aggregation layers. Also, the performance improvement of the proposed method compared with Meta-GNN might come from the usage of metric learning.

Meta-learning has received a tremendous recent attention as a possible approach for mimicking human intelligence, i.e., acquiring new knowledge and skills with little or even no demonstration. Most of the existing meta-learning methods are proposed to tackle few-shot learning problems such as image and text, in rather Euclidean domain. However, there are very few works applying meta-learning to non-Euclidean domains, and the recently proposed graph neural networks (GNNs) models do not perform effectively on graph few-shot learning problems. Towards this, we propose a novel graph meta-learning framework -- Meta-GNN -- to tackle the few-shot node classification problem in graph meta-learning settings. It obtains the prior knowledge of classifiers by training on many similar few-shot learning tasks and then classifies the nodes from new classes with only few labeled samples. Additionally, Meta-GNN is a general model that can be straightforwardly incorporated into any existing state-of-the-art GNN. Our experiments conducted on three benchmark datasets demonstrate that our proposed approach not only improves the node classification performance by a large margin on few-shot learning problems in meta-learning paradigm, but also learns a more general and flexible model for task adaption.