Active Semi-Supervised Learning Using Sampling Theory for Graph Signals

被引:93
|
作者
Gadde, Akshay [1 ]
Anis, Aamir [1 ]
Ortega, Antonio [1 ]
机构
[1] Univ Southern Calif, Dept Elect Engn, Los Angeles, CA 90007 USA
关键词
Active semi-supervised learning; Graph signal processing; Sampling theory; Graph signal filtering; REGULARIZATION; RECONSTRUCTION; FRAMEWORK;
D O I
10.1145/2623330.2623760
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We consider the problem of offline, pool-based active semi-supervised learning on graphs. This problem is important when the labeled data is scarce and expensive whereas unlabeled data is easily available. The data points are represented by the vertices of an undirected graph with the similarity between them captured by the edge weights. Given a target number of nodes to label, the goal is to choose those nodes that are most informative and then predict the unknown labels. We propose a novel framework for this problem based on our recent results on sampling theory for graph signals. A graph signal is a real-valued function defined on each node of the graph. A notion of frequency for such signals can be defined using the spectrum of the graph Laplacian matrix. The sampling theory for graph signals aims to extend the traditional Nyquist-Shannon sampling theory by allowing us to identify the class of graph signals that can be reconstructed from their values on a subset of vertices. This approach allows us to define a criterion for active learning based on sampling set selection which aims at maximizing the frequency of the signals that can be reconstructed from their samples on the set. Experiments show the effectiveness of our method.
引用
收藏
页码:492 / 501
页数:10
相关论文
共 50 条
  • [1] A Sampling Theory Perspective of Graph-Based Semi-Supervised Learning
    Anis, Aamir
    El Gamal, Aly
    Avestimehr, A. Salman
    Ortega, Antonio
    [J]. IEEE TRANSACTIONS ON INFORMATION THEORY, 2019, 65 (04) : 2322 - 2342
  • [2] Graph based semi-supervised learning using spatial segregation theory
    Bozorgnia, Farid
    Fotouhi, Morteza
    Arakelyan, Avetik
    Elmoataz, Abderrahim
    [J]. JOURNAL OF COMPUTATIONAL SCIENCE, 2023, 74
  • [3] SemiSegPolyp: Semi-Supervised Polyp Segmentation using Graph Signals
    El Rai, Marwa Chendeb
    Darweesh, Muna
    Al-Saad, Mina
    Mansoor, Wathiq
    Al-Ahmad, Hussain
    [J]. 2022 29TH IEEE INTERNATIONAL CONFERENCE ON ELECTRONICS, CIRCUITS AND SYSTEMS (IEEE ICECS 2022), 2022,
  • [4] ASYMPTOTIC JUSTIFICATION OF BANDLIMITED INTERPOLATION OF GRAPH SIGNALS FOR SEMI-SUPERVISED LEARNING
    Anis, Aamir
    El Gamal, Aly
    Avestimehr, Salman
    Ortega, Antonio
    [J]. 2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP), 2015, : 5461 - 5465
  • [5] Negative sampling in semi-supervised learning
    Chen, John
    Shah, Vatsal
    Kyrillidis, Anastasios
    [J]. 25TH AMERICAS CONFERENCE ON INFORMATION SYSTEMS (AMCIS 2019), 2019,
  • [6] Graph Construction for Semi-Supervised Learning
    Berton, Lilian
    Lopes, Alneu de Andrade
    [J]. PROCEEDINGS OF THE TWENTY-FOURTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI), 2015, : 4343 - 4344
  • [7] Efficiently Learning the Graph for Semi-supervised Learning
    Sharma, Dravyansh
    Jones, Maxwell
    [J]. UNCERTAINTY IN ARTIFICIAL INTELLIGENCE, 2023, 216 : 1900 - 1910
  • [8] Negative sampling in semi-supervised learning
    Chen, John
    Shah, Vatsal
    Kyrillidis, Anastasios
    [J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 119, 2020, 119
  • [10] Model Change Active Learning in Graph-Based Semi-supervised Learning
    Miller, Kevin S.
    Bertozzi, Andrea L.
    [J]. COMMUNICATIONS ON APPLIED MATHEMATICS AND COMPUTATION, 2024, 6 (02) : 1270 - 1298