Graph convolutional networks (GCNs) have recently received wide attentions, due to their successful applications in different graph tasks and different domains. Training GCNs for a large graph, however, is still a challenge. Original full-batch GCN training requires calculating the representation of all the nodes in the graph per GCN layer, which brings in high computation and memory costs. To alleviate this issue, several sampling-based methods have been proposed to train GCNs on a subset of nodes. Among them, the node-wise neighbor-sampling method recursively samples a fixed number of neighbor nodes, and thus its computation cost suffers from exponential growing neighbor size; while the layer-wise importance-sampling method discards the neighbor-dependent constraints, and thus the nodes sampled across layer suffer from sparse connection problem. To deal with the above two problems, we propose a new effective sampling algorithm called LAyer-Dependent ImportancE Sampling (LADIES) 2 . Based on the sampled nodes in the upper layer, LADIES selects their neighborhood nodes, constructs a bipartite subgraph and computes the importance probability accordingly. Then, it samples a fixed number of nodes by the calculated probability, and recursively conducts such procedure per layer to construct the whole computation graph. We prove theoretically and experimentally, that our proposed sampling algorithm outperforms the previous sampling methods in terms of both time and memory costs. Furthermore, LADIES is shown to have better generalization accuracy than original full-batch GCN, due to its stochastic nature. * equal contribution 2 codes are avaiable at https://github.com/acbull/LADIES 3 https://zephoria.com/top-15-valuable-facebook-statistics/
Anastomotic techniques are of vital importance in restoring gastrointestinal continuity after resection. An alternative asymmetric figure-of-eight single-layer suture anastomotic technique was introduced and its effects were evaluated in an in vitro porcine model. Twelve 15-cm grossly healthy small intestine segments from a porcine cadaver were harvested and randomly divided into asymmetric figure-of-eight single-layer suture (figure-of-eight suture) and single-layer interrupted suture technique (interrupted suture) groups (n = 6 in each group). The anastomosed bowel was infused with methylene blue solution to test anastomotic leakage. Anastomosis construction time, leakage, and suture material cost were recorded and analyzed statistically using Fisher's exact test and Student's t-test. One anastomotic leakage occurred (16.67%) in the figure-of-eight suture group, and two (33.33%) in the interrupted suture group (p > 0.9999). The anastomosis construction time was relatively short in the figure-of-eight suture group, but the difference did not reach a statistically significant level between the two groups. The mean number of suture knots and the cost of suture material in the figure-of-eight suture group were significantly decreased in comparison to the interrupted suture group (15.67 ± 3.30 vs. 22.17 ± 2.03, 167.11 ± 35.20 vs. 236.45 ± 21.70 CNY, p < 0.01, respectively). Our results suggested that the alternative asymmetric figure-of-eight suture technique was safe and economic for intestinal anastomosis. An in vivo experiment is required to elucidate the effects of this suture technique on the physiological anastomotic healing process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.