Filters








18,494 Hits in 2.6 sec

Learning to Propagate for Graph Meta-Learning [article]

Lu Liu, Tianyi Zhou, Guodong Long, Jing Jiang, Chengqi Zhang
2019 arXiv   pre-print
The meta-learner, called "Gated Propagation Network (GPN)", learns to propagate messages between prototypes of different classes on the graph, so that learning the prototype of each class benefits from  ...  Meta-learning extracts common knowledge from learning different tasks and uses it for unseen tasks.  ...  Government through the Australian Research Council (ARC) under grants 1) LP160100630 partnership with Australia Government Department of Health and 2) LP150100671 partnership with Australia Research Alliance for  ... 
arXiv:1909.05024v2 fatcat:gz7zoy244bdj3ju3l3zi6h3yue

Meta Propagation Networks for Graph Few-shot Semi-supervised Learning [article]

Kaize Ding, Jianling Wang, James Caverlee, Huan Liu
2022 arXiv   pre-print
Inspired by the extensive success of deep learning, graph neural networks (GNNs) have been proposed to learn expressive node representations and demonstrated promising performance in various graph learning  ...  In essence, our framework Meta-PN infers high-quality pseudo labels on unlabeled nodes via a meta-learned label propagation strategy, which effectively augments the scarce labeled data while enabling large  ...  To address the aforementioned challenges, we propose a new graph meta-learning framework, Meta Propagation Networks (Meta-PN), which goes beyond the canonical messagepassing scheme of GNNs and learns expressive  ... 
arXiv:2112.09810v2 fatcat:723hwqhbaff55lqifrnscmh6ay

Semi-supervised Learning over Heterogeneous Information Networks by Ensemble of Meta-graph Guided Random Walks

He Jiang, Yangqiu Song, Chenguang Wang, Ming Zhang, Yizhou Sun
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
After we obtain all the labels propagated by different trials of random walk guided by meta-graphs, we use an ensemble algorithm to vote for the final labeling results.  ...  Then we perform random walk over the sub-graphs to propagate the labels from labeled data to unlabeled data.  ...  Then it jointly learns the propagated labels and the weights for different meta-paths to propagate the labels.  ... 
doi:10.24963/ijcai.2017/270 dblp:conf/ijcai/JiangSWZS17 fatcat:6bcmpzl5lvdtvig3aoejtxjism

Prototype Propagation Networks (PPN) for Weakly-supervised Few-shot Learning on Category Graph [article]

Lu Liu, Tianyi Zhou, Guodong Long, Jing Jiang, Lina Yao, Chengqi Zhang
2019 arXiv   pre-print
Given a category graph of the targeted fine-classes and some weakly-labeled coarse-classes, PPN learns an attention mechanism which propagates the prototype of one class to another on the graph, so that  ...  The resulting graph of prototypes can be continually re-used and updated for new tasks and classes.  ...  We propose a meta-learning model called prototype propagation network (PPN) to explore the above weaklysupervised information for few-shot classification tasks.  ... 
arXiv:1905.04042v2 fatcat:t35txir2uvdjbawhqrpwbbu5j4

Looking Back to Lower-Level Information in Few-Shot Learning

Zhongjie Yu, Sebastian Raschka
2020 Information  
In particular, we propose the Looking-Back method, which could use lower-level information to construct additional graphs for label propagation in limited data settings.  ...  The novel contribution of this paper is the utilization of lower-level information to improve the meta-learner performance in few-shot learning.  ...  To test this hypothesis, we adopt the widely used Conv-64F [30] in few-shot learning as a backbone, and construct graphs for label propagation, following the transductive propagation network (TPN) [  ... 
doi:10.3390/info11070345 fatcat:jnryxvo56ffvza4kgw6telqu3m

Attribute Propagation Network for Graph Zero-Shot Learning

Lu Liu, Tianyi Zhou, Guodong Long, Jing Jiang, Chengqi Zhang
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
For better generalization over unseen classes, different from previous methods, we adopt a meta-learning strategy to train the propagation mechanism and the similarity metric for the NN classifier on multiple  ...  When the graph is not provided, given pre-defined semantic embeddings of the classes, we can learn a mechanism to generate the graph in an end-to-end manner along with the propagation mechanism.  ...  In the rest of this section, we illustrate solutions to four of these challenges. 1. How to choose the nodes for the propagation graph? 2. How to represent the node features in the propagation graph?  ... 
doi:10.1609/aaai.v34i04.5923 fatcat:bfm2uwocnbgjdcz36ucmuuokfa

TGG: Transferable Graph Generation for Zero-shot and Few-shot Learning [article]

Chenrui Zhang, Xiaoqing Lyu, Zhi Tang
2019 arXiv   pre-print
The two propagations learn from each other in a dual learning fashion, which performs as an adaptation way for mitigating domain shift.  ...  All components are jointly optimized with a meta-learning strategy, and our TGG acts as an end-to-end framework unifying conventional zero-shot, generalized zero-shot and few-shot learning.  ...  Relation propagation module exploits generated relation graph for classification, via a dual relation propagation approach with meta-learning strategy. 3.1.2 Class-level graph construction.  ... 
arXiv:1908.11503v1 fatcat:sqqgujd2fbbffduosccf3muzba

Learning Augmentation for GNNs with Consistency Regularization

Hyeonjin Park, Seunghun Lee, Dasol Hwang, Jisu Jeong, Kyung-Min Kim, Jung-Woo Ha, Hyunwoo J. Kim
2021 IEEE Access  
For instance, Meta-GNN [42] and Meta-Graph [43] use a gradient-based meta-learning method in a few-shot setting for node classification and link prediction, respectively.  ...  MetaR [44] is a meta-learning framework leveraging a knowledge graph for few-shot link prediction.  ... 
doi:10.1109/access.2021.3111908 fatcat:u5jodtmlbvhr7l3ttzbc2hl5lu

Contrastive Meta Learning with Behavior Multiplicity for Recommendation [article]

Wei Wei and Chao Huang and Lianghao Xia and Yong Xu and Jiashu Zhao and Dawei Yin
2022 arXiv   pre-print
To tackle the above challenges, we devise a new model CML, Contrastive Meta Learning (CML), to maintain dedicated cross-type behavior dependency for different users.  ...  Our empirical studies further suggest that the contrastive meta learning paradigm offers great potential for capturing the behavior multiplicity in recommendation.  ...  For all graphbased baselines, the number of graph-based message propagation layers is tuned from {1,2,3,4}.  ... 
arXiv:2202.08523v1 fatcat:mvx3u5hxkvh2hekxisptmupyp4

Semi-Stacking for Semi-supervised Sentiment Classification

Shoushan Li, Lei Huang, Jingjing Wang, Guodong Zhou
2015 Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)  
the meta-classifier.  ...  Specifically, we apply metalearning to predict the unlabeled data given the outputs from the member algorithms and propose N-fold cross validation to guarantee a suitable size of the data for training  ...  Due to the limited number of labeled data in semi-supervised learning, we use Nfold cross validation to obtain more meta-samples for better learning the meta-classifier.  ... 
doi:10.3115/v1/p15-2005 dblp:conf/acl/LiHWZ15a fatcat:rztox6tbl5bann33u2poj2aqke

MetaKG: Meta-learning on Knowledge Graph for Cold-start Recommendation [article]

Yuntao Du, Xinjun Zhu, Lu Chen, Ziquan Fang, Yunjun Gao
2022 arXiv   pre-print
The collaborative-aware meta learner aims to locally aggregate user preferences for each user preference learning task.  ...  meta learner, to capture meta users' preference and entities' knowledge for cold-start recommendations.  ...  Graph Neural Networks (GNNs) Given a graph consisting of nodes and edges, GNNs aim to learn the representation for each node by leveraging the information propagation from its neighborhoods in both spectral  ... 
arXiv:2202.03851v1 fatcat:ue5s7yvno5bnzi5mj5widt2obq

Jittor: a novel deep learning framework with meta-operators and unified graph execution

Shi-Min Hu, Dun Liang, Guo-Ye Yang, Guo-Wei Yang, Wen-Yang Zhou
2020 Science China Information Sciences  
Jittor: a novel deep learning framework with meta-operators and unified graph execution. Sci China Inf Sci, 2020, 63(12): 222103, https://doi.  ...  This approach is as easy to use as dynamic graph execution yet has the efficiency of static graph execution.  ...  We would like to thank anonymous reviewers for their helpful review comments, and Professor Ralph R. MARTIN for his useful suggestions and great help in paper writting.  ... 
doi:10.1007/s11432-020-3097-4 fatcat:t4ruztzhgbap5gvm4cliago5de

GCN for HIN via Implicit Utilization of Attention and Meta-paths [article]

Di Jin, Zhizhi Yu, Dongxiao He, Carl Yang, Philip S. Yu, Jiawei Han
2020 arXiv   pre-print
Graph neural networks for HIN embeddings typically adopt a hierarchical attention (including node-level and meta-path-level attentions) to capture the information from meta-path-based neighbors.  ...  layer-by-layer, realizing the function of attentions for selecting meta-paths in an indirect way.  ...  We use the direct linked meta-paths, a discriminative aggregation, along with the stacked layers of propagation, to distinguish the importance of different meta-paths.  ... 
arXiv:2007.02643v1 fatcat:k3jmh4z7lnd3fdshuvx6g4bude

DPGN: Distribution Propagation Graph Network for Few-shot Learning [article]

Ling Yang, Liangliang Li, Zilun Zhang, Xinyu Zhou, Erjin Zhou, Yu Liu
2020 arXiv   pre-print
We propose a novel approach named distribution propagation graph network (DPGN) for few-shot learning.  ...  Most graph-network-based meta-learning approaches model instance-level relation of examples.  ...  Our main contributions are summarized as follows: • To the best of our knowledge, DPGN is the first to explicitly incorporate distribution propagation in graph network for few-shot learning.  ... 
arXiv:2003.14247v2 fatcat:aucvyiowirbubezzprpyfpxb5e

Few-Shot Object Detection via Knowledge Transfer [article]

Geonuk Kim, Hong-Gyu Jung, Seong-Whan Lee
2020 arXiv   pre-print
To facilitate the remodeling process, we predict the prototypes under a graph structure, which propagates information of the correlated base categories to the novel categories with explicit guidance of  ...  Central to our method is prototypical knowledge transfer with an attached meta-learner.  ...  With a constructed meta-graph, we prepare to use a graph propagation mechanism to transfer node messages through the graph.  ... 
arXiv:2008.12496v1 fatcat:fwjlahzpu5fjlghskaj5qqu7gm
« Previous Showing results 1 — 15 out of 18,494 results