Learning from Dark : Boosting Graph Convolutional Neural Networks with Diverse Negative Samples

Conference paper


Duan, Wei, Xuan, Junyu, Qiao, Maoying and Lu, Jie. (2022). Learning from Dark : Boosting Graph Convolutional Neural Networks with Diverse Negative Samples. Thirty-Sixth AAAI Conference on Artificial Intelligence. 22 Feb - 01 Mar 2022 Canada: Association for the Advancement of Artificial Intelligence (AAAI). pp. 6650-6658
AuthorsDuan, Wei, Xuan, Junyu, Qiao, Maoying and Lu, Jie
TypeConference paper
Abstract

Graph Convolutional Neural Networks (GCNs) have been generally accepted to be an effective tool for node representations learning. An interesting way to understand GCNs is to think of them as a message passing mechanism where each node updates its representation by accepting information from its neighbours (also known as positive samples). However, beyond these neighbouring nodes, graphs have a large, dark, all-but forgotten world in which we find the non-neighbouring nodes (negative samples). In this paper, we show that this great dark world holds a substantial amount of information that might be useful for representation learning. Most specifically, it can provide negative information about the node representations. Our overall idea is to select appropriate negative samples for each node and incorporate the negative information contained in these samples into the representation updates. Moreover, we show that the process of selecting the negative samples is not trivial. Our theme therefore begins by describing the criteria for a good negative sample, followed by a determinantal point process algorithm for efficiently obtaining such samples. A GCN, boosted by diverse negative samples, then jointly considers the positive and negative information when passing messages. Experimental evaluations show that this idea not only improves the overall performance of standard representation learning but also significantly alleviates over-smoothing problems.

KeywordsMachine Learning; ML; Knowledge Representation And Reasoning; KRR
Year01 Jan 2022
PublisherAssociation for the Advancement of Artificial Intelligence (AAAI)
ISSN2374-3468
Web address (URL)https://ojs.aaai.org/index.php/AAAI/article/view/20608
Open accessOpen access
Research or scholarlyResearch
Publisher's version
License
All rights reserved
File Access Level
Open
Page range6650-6658
ISBN1-57735-876-7
Web address (URL) of conference proceedingshttps://ojs.aaai.org/index.php/AAAI/issue/view/512
Output statusPublished
Publication dates
Online30 Jun 2022
Publication process dates
Completed01 Mar 2022
Deposited06 Feb 2024
Additional information

Copyright © 2022, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.

Place of publicationCanada
Permalink -

https://acuresearchbank.acu.edu.au/item/901wq/learning-from-dark-boosting-graph-convolutional-neural-networks-with-diverse-negative-samples

Download files


Publisher's version
Qiao_2022_Learning_from_dark_boosting_graph_convolutional.pdf
License: All rights reserved
File access level: Open

  • 14
    total views
  • 11
    total downloads
  • 3
    views this month
  • 4
    downloads this month
These values are for the period from 19th October 2020, when this repository was created.

Export as

Related outputs

Deep learning methods applied to electronic monitoring data : Automated catch event detection for longline fishing
Qiao, Maoying, Wang, Dadong, Tuck, Geoffrey N., Little, L. Richard, Punt, Andre E. and Gerner, Mike. (2021). Deep learning methods applied to electronic monitoring data : Automated catch event detection for longline fishing. ICES Journal of Marine Science: journal du conseil. 78(1), pp. 25-35. https://doi.org/10.1093/icesjms/fsaa158
Diversified Bayesian nonnegative matrix factorization
Qiao, Maoying, Jun,Yu, Tongliang, Liu, Xinchao, Wang and Dacheng, Tao. (2020). Diversified Bayesian nonnegative matrix factorization. The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20). New York Hilton Midtown, New York, New York, United States of America 07 - 12 Feb 2020 AAAI Press. pp. 5420-5427 https://doi.org/10.1609/aaai.v34i04.5991
Adapting stochastic block models to power-law degree distributions
Qiao, Maoying, Yu, Jun, Bian, Wei, Li, Qiang and Tao, Dacheng. (2019). Adapting stochastic block models to power-law degree distributions. IEEE Transactions on Cybernetics. 49(2), pp. 626-637. https://doi.org/10.1109/TCYB.2017.2783325Y
Diversified dictionaries for multi-instance learning
Qiao, Maoying, Liu, Liu, Yu, Jun, Xu, Chang and Tao, Dacheng. (2017). Diversified dictionaries for multi-instance learning. Pattern Recognition. 64, pp. 407-416. https://doi.org/10.1016/j.patcog.2016.08.026
Improving stochastic block models by incorporating power-law degree characteristic
Qiao, Maoying, Yu, Jun, Bian, Wei, Li, Qiang and Tao, Dacheng. (2017). Improving stochastic block models by incorporating power-law degree characteristic. Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17). Melbourne, Australia 19 - 25 Aug 2017 International Joint Conferences on Artificial Intelligence Organization. pp. 2620-2626 https://doi.org/10.24963/ijcai.2017/365
Fast sampling for time-varying determinantal point processes
Qiao, Maoying, Xu, Richard Yi Da, Bian, Wei and Tao, Dacheng. (2016). Fast sampling for time-varying determinantal point processes. ACM Transactions on Knowledge Discovery from Data. 11(1), p. 8. https://doi.org/1556-4681
Conditional graphical lasso for multi-label image classification
Li, Qiang, Qiao, Maoying, Bian, Wei and Tao, Dacheng. (2016). Conditional graphical lasso for multi-label image classification. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, Nevada, United States of America 27 - 30 Jun 2016 Computer Vision Foundation. pp. 2977-2986 https://doi.org/10.1109/CVPR.2016.325
Diversified hidden Markov models for sequential labeling
Qiao, Maoying, Bian, Wei, Da Xu, Richard Yi and Tao, Dacheng. (2015). Diversified hidden Markov models for sequential labeling. IEEE Transactions on Knowledge and Data Engineering. 27(11), pp. 2947-2960. https://doi.org/10.1109/TKDE.2015.2433262
Biview learning for human posture segmentation from 3D points cloud
Qiao, Maoying, Cheng, Jun, Bian, Wei and Tao, Dacheng. (2014). Biview learning for human posture segmentation from 3D points cloud. PLoS ONE. 9(1), p. e85811. https://doi.org/10.1371/journal.pone.0085811