The problem of session-based recommendation aims to predict user actions based on anonymous sessions. Previous methods model a session as a sequence and estimate user representations besides item representations to make recommendations. Though achieved promising results, they are insufficient to obtain accurate user vectors in sessions and neglect complex transitions of items. To obtain accurate item embedding and take complex transitions of items into account, we propose a novel method, i.e. Session-based Recommendation with Graph Neural Networks, SR-GNN for brevity. In the proposed method, session sequences are modeled as graphstructured data. Based on the session graph, GNN can capture complex transitions of items, which are difficult to be revealed by previous conventional sequential methods. Each session is then represented as the composition of the global preference and the current interest of that session using an attention network. Extensive experiments conducted on two real datasets show that SR-GNN evidently outperforms the state-of-the-art session-based recommendation methods consistently.

No abstract

Graph representation learning nowadays becomes fundamental in analyzing graphstructured data. Inspired by recent success of contrastive methods, in this paper, we propose a novel framework for unsupervised graph representation learning by leveraging a contrastive objective at the node level. Specifically, we generate two graph views by corruption and learn node representations by maximizing the agreement of node representations in these two views. To provide diverse node contexts for the contrastive objective, we propose a hybrid scheme for generating graph views on both structure and attribute levels. Besides, we provide theoretical justification behind our motivation from two perspectives, mutual information and the classical triplet loss. We perform empirical experiments on both transductive and inductive learning tasks using a variety of real-world datasets. Experimental experiments demonstrate that despite its simplicity, our proposed method consistently outperforms existing state-of-the-art methods by large margins. Notably, our method gains about 10% absolute improvements on protein function prediction. Our unsupervised method even surpasses its supervised counterparts on transductive tasks, demonstrating its great potential in real-world applications.Recently, graph representation learning using Graph Neural Networks (GNN) has received considerable attention. Along with its prosperous development, however, there is an increasing concern over the label availability when training the model. Nevertheless, existing GNN models are mostly established in a supervised manner [6][7][8], which require abundant labeled nodes for training. Albeit with some attempts connecting previous unsupervised objectives (i.e., matrix reconstruction) to GNN models [9, 10], these methods still heavily rely on the preset graph proximity matrix.

The E4 allele of the Apolipoprotein E (APOE) gene is the strongest genetic risk factor for late-onset Alzheimer's disease (AD), and affects clinical outcomes of chronic and acute brain damages. The mechanisms by which apoE affect diverse diseases and disorders may involve modulation of the glial response to various types of brain damages. We examined glial activation in a mouse model where each of the human APOE alleles are expressed under the endogenous mouse APOE promoter, as well as in APOE knock-out mice. APOE4 mice displayed increased glial activation in response to intracerebroventricular lipopolysaccharide (LPS) compared to APOE2 and APOE3 mice by several measures. There were higher levels of microglia/macrophage, astrocytes, and invading T-cells after LPS injection in APOE4 mice. APOE4 mice also displayed greater and more prolonged increases of cytokines (IL-1β, IL-6, TNF-α) than APOE2 and APOE3 mice. We found that APOE4 mice had greater synaptic protein loss after LPS injection, as measured by three different markers: PSD-95, Drebin, and synaptophysin. In all assays, APOE knock-out mice responded similar to APOE4 mice, suggesting that the apoE4 protein may lack anti-inflammatory characteristics of apoE2 and apoE3. Together, these findings demonstrate that APOE4 predisposes to inflammation, which could contribute to its association with Alzheimer's disease and other disorders.