Filters








58,457 Hits in 3.7 sec

Unsupervised Dependency Graph Network

Yikang Shen, Shawn Tan, Alessandro Sordoni, Peng Li, Jie Zhou, Aaron Courville
2022 Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)   unpublished
We introduce a new model, the Unsupervised Dependency Graph Network (UDGN), that can induce dependency structures from raw corpora and the masked language modeling task.  ...  Experiment results show that UDGN achieves very strong unsupervised dependency parsing performance without gold POS tags and any other external information.  ...  Building on these components, we propose a novel architecture, the Unsupervised Dependency Graph Network (UDGN).  ... 
doi:10.18653/v1/2022.acl-long.327 fatcat:cephsaawxndpbawvjagwpvw54y

GLoMo: Unsupervised Learning of Transferable Relational Graphs

Zhilin Yang, Junbo Jake Zhao, Bhuwan Dhingra, Kaiming He, William W. Cohen, Ruslan Salakhutdinov, Yann LeCun
2018 Neural Information Processing Systems  
This work explores the possibility of learning generic latent relational graphs that capture dependencies between pairs of data units (e.g., words or pixels) from large-scale unlabeled data and transferring  ...  the graphs to downstream tasks.  ...  During the unsupervised learning phase, our framework trains two networks, a graph predictor network g and a feature predictor network f .  ... 
dblp:conf/nips/YangZDHCSL18 fatcat:4pviuw5guzgg7lu5i7y2llstei

Unsupervised Hierarchical Graph Representation Learning by Mutual Information Maximization [article]

Fei Ding, Xiaohong Zhang, Justin Sybrandt, Ilya Safro
2020 arXiv   pre-print
Graph representation learning based on graph neural networks (GNNs) can greatly improve the performance of downstream tasks, such as node and graph classification.  ...  To address these issues, we present an unsupervised graph representation learning method, Unsupervised Hierarchical Graph Representation (UHGR), which can generate hierarchical representations of graphs  ...  In this work, we consider two general GNNs: Graph Convolutional Networks (GCNs) [19] and Graph Attention Networks (GATs) [38] . Graph Convolutional Networks.  ... 
arXiv:2003.08420v3 fatcat:n2u3becfmzdkfmyei7q3qm5xoi

Study on a New Method of Link-Based Link Prediction in the Context of Big Data

Chen Jicheng, Chen Hongchang, Li Hanchao, Fahd Abd Algalil
2021 Applied Bionics and Biomechanics  
The resolutions made by addressing the above techniques place our framework above the previous literature's unsupervised approaches.  ...  A barrage of literature has been written to approach this problem; however, they mostly come from the angle of unsupervised learning (UL).  ...  dependent on the network's average interaction frequency (iii) Graph Weighting.  ... 
doi:10.1155/2021/1654134 fatcat:lzra62guyfdf7nkxkcfbqvujbi

GLoMo: Unsupervisedly Learned Relational Graphs as Transferable Representations [article]

Zhilin Yang, Jake Zhao, Bhuwan Dhingra, Kaiming He, William W. Cohen, Ruslan Salakhutdinov, Yann LeCun
2018 arXiv   pre-print
This work explores the possibility of learning generic latent relational graphs that capture dependencies between pairs of data units (e.g., words or pixels) from large-scale unlabeled data and transferring  ...  the graphs to downstream tasks.  ...  During the unsupervised learning phase, our framework trains two networks, a graph predictor network g and a feature predictor network f .  ... 
arXiv:1806.05662v3 fatcat:ype6zmg4jvdhzp5owcyewfgrde

StructFormer: Joint Unsupervised Induction of Dependency and Constituency Structure from Masked Language Modeling [article]

Yikang Shen, Yi Tay, Che Zheng, Dara Bahri, Donald Metzler, Aaron Courville
2021 arXiv   pre-print
To achieve this, we propose a new parsing framework that can jointly generate a constituency tree and dependency graph.  ...  Experimental results show that our model can achieve strong results on unsupervised constituency parsing, unsupervised dependency parsing, and masked language modeling at the same time.  ...  Unsupervised Dependency Parsing The unsupervised dependency parsing evaluation compares the induced dependency relations with those in the reference dependency graph.  ... 
arXiv:2012.00857v3 fatcat:ph3pvht7gvenzdnr7lcftvlegm

Wireless Link Scheduling via Graph Representation Learning: A Comparative Study of Different Supervision Levels [article]

Navid Naderializadeh
2021 arXiv   pre-print
We leverage the interference graph of the wireless network as an underlying topology for a graph neural network (GNN) backbone, which converts the channel matrix to a set of node embeddings for all transmitter-receiver  ...  We consider the problem of binary power control, or link scheduling, in wireless interference networks, where the power control policy is trained using graph representation learning.  ...  In the following, we drop the dependence of the graph G on H for brevity unless necessary.  ... 
arXiv:2110.01722v1 fatcat:7hwv2vvv2bcx7m2ilisioex3qu

Unsupervised Graph Representation by Periphery and Hierarchical Information Maximization [article]

Sambaran Bandyopadhyay, Manasvi Aggarwal, M. Narasimha Murty
2020 arXiv   pre-print
Thus, we aim to propose an unsupervised graph neural network to generate a vector representation of an entire graph in this paper.  ...  However, for the entire graph representation, most of the existing graph neural networks are trained on a graph classification loss in a supervised way.  ...  One limitation of most of the existing graph neural networks for graph level task is their dependency on the availability of graph labels.  ... 
arXiv:2006.04696v1 fatcat:ihbpx22jfvhx5pzhajeznyytyq

Reconstructive Sequence-Graph Network for Video Summarization

Bin Zhao, Haopeng Li, Xiaoqiang Lu, Xuelong Li
2021 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Long Short-Term Memory (LSTM), and the shot-level dependencies are captured by the Graph Convolutional Network (GCN).  ...  Motivated by this point, we propose a Reconstructive Sequence-Graph Network (RSGN) to encode the frames and shots as sequence and graph hierarchically, where the frame-level dependencies are encoded by  ...  However, our approach captures the global dependencies among shots using graph convolution network (GCN) rather than LSTM.  ... 
doi:10.1109/tpami.2021.3072117 pmid:33835915 fatcat:x5ql4fld5barnndnv2uizejeau

PyGOD: A Python Library for Graph Outlier Detection [article]

Kay Liu, Yingtong Dou, Yue Zhao, Xueying Ding, Xiyang Hu, Ruitong Zhang, Kaize Ding, Canyu Chen, Hao Peng, Kai Shu, George H. Chen, Zhihao Jia (+1 others)
2022 arXiv   pre-print
PyGOD is an open-source Python library for detecting outliers on graph data.  ...  As the first comprehensive library of its kind, PyGOD supports a wide array of leading graph-based methods for node-, edge-, subgraph-, and graph-level outlier detection, under a unified, well-documented  ...  Library Design and Implementation Dependency.  ... 
arXiv:2204.12095v1 fatcat:j6jabdwa4vcgvo4yxzzrun73f4

GAR: An efficient and scalable Graph-based Activity Regularization for semi-supervised learning [article]

Ozsel Kilinc, Ismail Uysal
2018 arXiv   pre-print
In this paper, we propose a novel graph-based approach for semi-supervised learning problems, which considers an adaptive adjacency of the examples throughout the unsupervised portion of the training.  ...  Ultimately, the proposed framework provides an effective and scalable graph-based solution which is natural to the operational mechanism of deep neural networks.  ...  G N : N = m /nI (7) Depending on this relation, we propose to apply regularization during the unsupervised task in order to ensure that N becomes the identity matrix.  ... 
arXiv:1705.07219v2 fatcat:wo25ralrafe5jkhxqnkhecyt74

A Survey of Unsupervised Dependency Parsing [article]

Wenjuan Han, Yong Jiang, Hwee Tou Ng, Kewei Tu
2020 arXiv   pre-print
Unsupervised dependency parsing aims to learn a dependency parser from sentences that have no annotation of their correct parse trees.  ...  In this paper, we survey existing approaches to unsupervised dependency parsing, identify two major classes of approaches, and discuss recent trends.  ...  The encoder of this model is a conditional random field model while the decoder generates a sentence based on a graph convolutional neural network whose structure is specified by the dependency tree.  ... 
arXiv:2010.01535v1 fatcat:4wd4dgducnbeti6kukpmfw5r4i

Discovering Multi-relational Latent Attributes by Visual Similarity Networks [chapter]

Fatemeh Shokrollahi Yancheshmeh, Joni-Kristian Kämäräinen, Ke Chen
2015 Lecture Notes in Computer Science  
Instead of clustering, a network (graph) containing multiple connections is a natural way to represent such multi-relational attributes between images.  ...  In the light of this, we introduce an unsupervised framework for network construction based on pairwise visual similarities and experimentally demonstrate that the constructed network can be used to automatically  ...  In a single network, there can be several almost equally good paths between two nodes and the transition "smoothness" depends on how many images there are in a network.  ... 
doi:10.1007/978-3-319-16634-6_1 fatcat:dlenin3kbfbbtl5mslrdvkv3fa

Anomaly Detection in Bitcoin Network Using Unsupervised Learning Methods [article]

Thai Pham, Steven Lee
2017 arXiv   pre-print
network: one graph has users as nodes, and the other has transactions as nodes.  ...  To this end, we use three unsupervised learning methods including k-means clustering, Mahalanobis distance, and Unsupervised Support Vector Machine (SVM) on two graphs generated by the Bitcoin transaction  ...  ., x m ) in which x i ∈ R n (where n = 6 or 3 depending on graphs) for each i = 1, ..., m.  ... 
arXiv:1611.03941v2 fatcat:ouk37gbmkbhrrbek3f25u5g45u

Residual or Gate? Towards Deeper Graph Neural Networks for Inductive Graph Representation Learning [article]

Binxuan Huang, Kathleen M. Carley
2019 arXiv   pre-print
We present a graph neural network class named recurrent graph neural network (RGNN), that address the shortcomings of prior methods.  ...  In this paper, we study the problem of node representation learning with graph neural networks.  ...  Li et al. use GRU to model the temporal dependency in a sequence of traffic network.  ... 
arXiv:1904.08035v3 fatcat:26wbxjpycfcahmnpz76sqse2ku
« Previous Showing results 1 — 15 out of 58,457 results