Graph representation learning has attracted tremendous attention due to its remarkable performance in variety of real-world applications. However, because data labeling is always time and resource intensive, current supervised graph representation learning models for particular tasks frequently suffer from label sparsity issues. In light of this, graph few-shot learning has been proposed to tackle the performance degradation in face of limited annotated data challenge. While recent advances in graph few shot learning achieve promising performance, they typically force to use a generic feature embedding across various tasks. Ideally, we want to construct feature embeddings that are tuned for the given task because of the differences in distribution between tasks. In this work, we propose a novel Task-Aware Graph Model (TAGM) to learn task-aware node embedding. Specifically, we provide a new graph cell design that includes a graph convolution layer for aggregating and updating graph information as well as a two-layer linear transformation for node feature transformation. On this basis, we encode task information to learn the binary weight mask set and gradient mask set, where the weight mask set selects different network parameters for different tasks and the gradient mask set can dynamically update the selected network parameters in a different manner during the optimization process. Our model is more sensitive to task identity and performs better for a task graph input. Our extensive experiments on three graph-structured datasets demonstrate that our proposed method generally outperforms the state-of-the-art baselines in few-shot learning.

Graph few-shot learning aims to learn how to quickly adapt to new tasks using only a few labeled data, which transfers learned knowledge of base classes to novel classes. Existing methods are mainly designed for static graphs, while many real-world graphs are dynamic and evolving over time, resulting in a phenomenon of structure and class evolutions. To address the challenges caused by the phenomenon, in this paper, we propose a novel algorithm named Learning to Generate Parameters (LGP) to deal with few-shot class-evolutionary learning on dynamic graphs. Specifically, for the structure evolution, LGP integrates ensemble learning into a backbone network to effectively learn invariant representation across different snapshots within a dynamic graph. For the class evolution, LGP adopts a meta-learning strategy that can learn to generate the classified parameters of novel classes via the parameters of the base classes. Therefore, LGP can quickly adapt to new tasks on a combination of base and novel classes. Besides, LGP utilizes an attention mechanism to capture the evolutionary pattern between the novel and based classes. Extensive experiments on a real-world dataset demonstrate the effectiveness of LGP.

The knowledge graph (KG) is an essential form of knowledge representation that has grown in prominence in recent years. Because it concentrates on nominal entities and their relationships, traditional knowledge graphs are static and encyclopedic in nature. On this basis, event knowledge graph (Event KG) models the temporal and spatial dynamics by text processing to facilitate downstream applications, such as question-answering, recommendation and intelligent search. Existing KG research, on the other hand, mostly focuses on text processing and static facts, ignoring the vast quantity of dynamic behavioral information included in photos, movies, and pre-trained neural networks. In addition, no effort has been done to include behavioral intelligence information into the knowedge graph for deep reinforcement learning (DRL) and robot learning. In this paper, we propose a noval dynamic knowledge and skill graph (KSG), and then we develop a basic and specific KSG based on CN-DBpedia. The nodes are divided into entity and attribute nodes, with entity nodes containing the agent, environment, and skill (DRL policy or policy representation), and attribute nodes containing the entity description, pre-train network, and offline dataset. KSG can search for different agents' skills in various environments and provide transferable information for acquiring new skills. This is the first study that we are aware of that looks into dynamic KSG for skill retrieval and learning. Extensive experimental results on new skill learning show that KSG boosts new skill learning efficiency.

In recent years, graph contrastive learning has achieved promising node classification accuracy using graph neural networks (GNNs), which can learn representations in an unsupervised manner. However, such representations cannot be generalized to unseen novel classes with only few-shot labeled samples in spite of exhibiting good performance on seen classes. In order to assign generalization capability to graph contrastive learning, we propose multimodal graph meta contrastive learning (MGMC) in this paper, which integrates multimodal meta learning into graph contrastive learning. On one hand, MGMC accomplishes effectively fast adapation on unseen novel classes by the aid of bilevel meta optimization to solve few-shot problems. On the other hand, MGMC can generalize quickly to a generic dataset with multimodal distribution by inducing the FiLM-based modulation module. In addition, MGMC incorporates the lastest graph contrastive learning method that does not rely on the onstruction of augmentations and negative examples. To our best knowledge, this is the first work to investigate graph contrastive learning for few-shot problems. Extensieve experimental results on three graph-structure datasets demonstrate the effectiveness of our proposed MGMC in few-shot node classification tasks.

Meta-learning aims to acquire common knowledge from a large amount of similar tasks and then adapts to unseen tasks within few gradient updates. Existing graph meta-learning algorithms show appealing performance in a variety of domains such as node classification and link prediction. These methods find a single common initialization for entire tasks and ignore the diversity of task distributions, which might be insufficient for multi-modal tasks. Recent approaches adopt modulation network to generate task-specific parameters for further achieving multiple initializations, which shows excellent performance for multi-modal image classification. However, different from image classification, how to design an effective modulation network to handle graph-structure dataset is still challenging. In this paper, we propose a Multi-Initialization Graph Meta-Learning (MI-GML) network for graph node classification, mainly consisting of local and global modulation neworks and meta learner. In terms of modulation network, we exploit local and global graph structure information to extract task-specific modulation parameters. On this basis, the meta learner is further modulated by the corresponding modulation parameter to produce task-specific representation for node classification. Experimental results on three graph-structure datasets demonstrate the effectiveness of MI-GML in few-shot node classification tasks.