A Graph is Worth 1-bit Spikes: When Graph Contrastive Learning Meets Spiking Neural Networks They propose a novel spiking graph contrastive learning framework to learn binarized 1-bit representations for graphs, making balanced trade-offs between efficiency and performance. Details Abstract: While cont...
A Graph is Worth 1-bit Spikes: When Graph Contrastive Learning Meets Spiking Neural Networks Jintang Li1, Huizhe Zhang1, Ruofan Wu2, Zulun Zhu3, Baokun Wang2, Changhua Meng2, Zibin Zheng1, Liang Chen1 1Sun Yat-sen University,2Ant Group,3Nanyang Technological University ...
生成式模型，如图自编码器（Graph Autoencoder, GAE），通过扰动原始数据来重构，旨在学习数据的内在结构。对比式模型，如图对比学习（Graph Contrastive Learning, GCL），通过最大化不同视图数据之间的信息熵来优化表示。GAE 方法自 1993 年由 Geoffrey Hinton 大师提出后，经历了从降噪自编码器（Denoising...
13、Graph contrastive learning automated. In ICML, 2021. 14、Graph contrastive learning with adaptive augmentation. In TheWebConf, 2021. 15、Graph information bottleneck for subgraph recognition. In ICLR, 2020. 16、Sub-graph contrast for scalable self-supervised graph representation learning. In ICDM...
对比式模型就是大名鼎鼎的对比学习(Contrastive Learning, CL),自监督的信号通常来自对原始图结构或属性的扰动,以最大化原始数据视图与自监督视图的互信息(Mutual Information, MI)作为学习目标。以图自编码器(Graph Autoencoder, GAE)为代表的生成式模型同样对原始数据进行扰动,但其学习目标是通过利用未被扰动的数据...
Zhang, C., et al.: When sparsity meets contrastive models: Less graph data can bring better class-balanced representations. In: ICML (2023) Google Scholar Zhang, C., Liu, H., Li, J., Ye, Y., Zhang, C.: Mind the gap: mitigating the distribution gap in graph few-shot learning. Tr...
Biswas D, Tešić J (2024) Domain adaptation with contrastive learning for object detection in satellite imagery. IEEE Trans Geosci Remote Sens Wang Y, Yue Z, Hua X-S, Zhang H (2023) Random boxes are open-world object detectors. In: Proceedings of the IEEE/CVF international conference on...
Inter-Modal Projector:A cross-modal alignment module like contrastive learning that maps graph and text vectors into a common embedding space. Language Decoder:An LLM like BERT that performs token-level reasoning on the fused graph-text representations from the projector. ...
learning PPIs. For example, two proteins with low scores of the “protein binding” function term hardly interact with each other. We suggest that future work may consider leveraging function annotations to enhance the expressiveness of protein representations. Inspired by the contrastive learning ...
A Graph is Worth 1-bit Spikes: When Graph Contrastive Learning Meets Spiking Neural Networks They propose a novel spiking graph contrastive learning framework to learn binarized 1-bit representations for graphs, making balanced trade-offs between efficiency and performance. Details Abstract: While cont...