Upload File

adversarial learning on heterogeneous information networks

  • Home
  • Documents
  • Adversarial Learning on Heterogeneous Information Networks
1
RESEARCH POSTER PRESENTATION DESIGN © 2015 www.PosterPresentations.com Background & Problem HeGAN : The Proposed Model 1 Beijing University of Posts and Telecommunications 2 AI Department, Ant Financial Services Group 3 Singapore Management University Binbin Hu 1, 2 , Yuan Fang 3 , Chuan Shi 1 Adversarial Learning on Heterogeneous Information Networks Heterogeneous Information Network (HIN) l Include multiple types of nodes and links l Model heterogeneous data and contain rich semantics HIN Embeddng with GAN based Adversarial Learning (HeGAN) Experiments Datasets C onclusions l We are the first to employ adversarial learning for HIN embedding, in order to utilize the rich semantics on HINs l We propose HeGAN that is not only relation-aware to capture rich semantics, but also equipped with a generalized generator l Extensive experimental results have revealed the effectiveness and efficiency of HeGAN Acknowledgements This research was supported by the National Natural Science Foun- dation of China (No. 61772082, 61702296), the National Key Research and Development Program of China (2017YFB0803304), the Beijing Municipal Natural Science Foundation (4182043) and the Singapore Ministry of Education (MOE) Academic Research Fund (AcRF) Tier 1 grant (Approval No. 18-C220-SMU-006). Contact l [email protected] l ([email protected]) l [email protected] l [email protected] HIN Embedding (HINE) l Consist of two samplers and one loss function l Samplers : select positive and negative examples l Loss function : trained on these samples to optimize node representations Adversarial Learning (or GAN) l Makes the model more robust to sparse or noisy data l Provides better samples to reduce the labeling requirement l GraphGAN, ANE, NETRA, ARGA Limitation of HINE l Randomly select existing nodes in the network as negative samples l Heed to the latent distribution of the nodes so that lack robustness l Require domain knowledge Limitation on GAN based Embedding l Only investigate homogeneous networks l Poor performance on semantic-rich HINs Our Idea HINE+ Adversarial Learning How to capture the semantics of multiple types of nodes and relations? How to generate fake samples efficiently and effectively? Relation-aware, generalized generator Relation-aware discriminator Calculate gradient for three cases I. Connected under given relation II. Connected under incorrect relation III. Fake node from relation-aware generator Challenges Relation-aware Generator and Discriminator I. Discriminator can tell whether a node pair is real or fake w.r.t relation II. Generator can produce fake node pairs that mimic real pairs w.r.t relation Solutions Generalized Generator I. Sample latent nodes from a continuous distribution II. No softmax computation and fake samples are not restricted to the existing nodes Node Clustering Node Classification Link Prediction Adversarial Learning Heterogeneity and Generalized Generator Efficiency Recommendation

Upload: others

Post on 18-Dec-2021

1 views

Category:

Documents


0 download

Report

TRANSCRIPT

Page 1: Adversarial Learning on Heterogeneous Information Networks

RESEARCH POSTER PRESENTATION DESIGN © 2015

www.PosterPresentations.com

Background & Problem

HeGAN : The Proposed Model

1Beijing University of Posts and Telecommunications2AI Department, Ant Financial Services Group

3Singapore Management University

Binbin Hu1, 2, Yuan Fang3, Chuan Shi1Adversarial Learning on Heterogeneous Information Networks

Heterogeneous Information Network (HIN)l Include multiple types of nodes and linksl Model heterogeneous data and contain rich semantics

HIN Embeddng with GAN based Adversarial Learning (HeGAN)

ExperimentsDatasets

Conclusionsl We are the first to employ adversarial learning for HIN

embedding, in order to utilize the rich semantics on HINsl We propose HeGAN that is not only relation-aware to capture

rich semantics, but also equipped with a generalized generatorl Extensive experimental results have revealed the effectiveness

and efficiency of HeGAN

AcknowledgementsThis research was supported by the National Natural Science Foun- dation ofChina (No. 61772082, 61702296), the National Key Research andDevelopment Program of China (2017YFB0803304), the Beijing MunicipalNatural Science Foundation (4182043) and the Singapore Ministry ofEducation (MOE) Academic Research Fund (AcRF) Tier 1 grant (ApprovalNo. 18-C220-SMU-006).

Contactl [email protected]

l ([email protected])

l [email protected]

l [email protected]

HIN Embedding (HINE)l Consist of two samplers and one loss functionl Samplers : select positive and negative examplesl Loss function : trained on these samples to optimize

node representations

Adversarial Learning (or GAN)l Makes the model more robust to sparse

or noisy datal Provides better samples to reduce the

labeling requirementl GraphGAN, ANE, NETRA, ARGA

Limitation of HINEl Randomly select existing nodes in the network as

negative samples l Heed to the latent distribution of the nodes so that

lack robustnessl Require domain knowledge

Limitation on GAN based Embeddingl Only investigate homogeneous networks l Poor performance on semantic-rich HINs

Our Idea

HINE+ Adversarial Learning

How to capture the semantics of multiple types of nodes and relations?

How to generate fake samples efficiently and effectively?

Relation-aware, generalized generator

Relation-aware discriminator

Calculate gradient for three cases

I. Connected under given relation

II. Connected under incorrect relation

III. Fake node from relation-aware generator

Challenges

Relation-aware Generator and DiscriminatorI. Discriminator can tell whether a node pair is real

or fake w.r.t relationII. Generator can produce fake node pairs that mimic

real pairs w.r.t relation

Solutions

Generalized GeneratorI. Sample latent nodes from a continuous

distributionII. No softmax computation and fake samples are

not restricted to the existing nodes

Node Clustering

Node Classification

Link Prediction

Adversarial Learning

Heterogeneity and Generalized Generator

EfficiencyRecommendation

Generative Adversarial Networksbhiksha/courses/deeplearning/Spring.2018/... · 2018. 3. 29. · Generative Adversarial Networks Generative adversarial networks (GANs) are relatively

heterogeneous sensor networks

Generative Adversarial Networks (part 2)deeplearning.cs.cmu.edu/document/slides/bstriner_gans_part_2.pdf · Unrolled Generative Adversarial Networks Unrolled Generative Adversarial

Generative Adversarial Networks and Their Applications

Generative Adversarial Networks for Simulation · Generative Adversarial Networks for Simulation! @wondermicky " [email protected] # mickypaganini.github.io Michela Paganini

Generative Adversarial Networks, and Applications Adversarial Networks, and Applications Outline: •Generative Models vs Discriminative Models (Background) •Generative Adversarial

Heterogeneous Data Networks

Generative adversarial networks

Generative Adversarial Networks (GANs) - Emanuele Sansone · List of recent papers on Generative Adversarial Networks: PAPERs about GANs 20142016 Generative Adversarial Networks ICLR

Adversarial Convolutional Networks with Weak Domain

COMP9444 Neural Networks and Deep Learning 9b. Generative Adversarial … · 2020. 7. 25. · COMP9444 20T2 Generative Adversarial Networks 13 Generative Adversarial Networks GAN

Flow Adversarial Networks: Flowrate Prediction for Gas ...yinyanliu.com/publications/Hu_Flow Adversarial Networks_2019.pdf · Flow Adversarial Networks: Flowrate Prediction for Gas–Liquid

Lec: Generative Adversarial Networks

Heterogeneous Cellular Networks

Generative Adversarial Networks for unsupervised monocular ...vision.deis.unibo.it/~smatt/Papers/3DRW2018/Monogan.pdf · Generative Adversarial Networks for unsupervised monocular

Geometry-Consistent Generative Adversarial Networks for ...openaccess.thecvf.com/content_CVPR_2019/papers/Fu_Geometry-Consistent... · Geometry-Consistent Generative Adversarial Networks

Adversarial Distillation of Bayesian Neural Network Posteriorsproceedings.mlr.press/v80/wang18i/wang18i.pdf · 2.3. Generative Adversarial Networks Generative Adversarial Networks

Lecture 7: Generative Adversarial Networks

Generative adversarial networks and convolutional …...Generative adversarial networks and convolutional neural networks based weather classification model for day ahead short-term

Generative adversarial networks...Generative adversarial nets. NeurIPS. Longer version Goodfellow, (2016). NIPS 2016 Tutorial: Generative Adversarial Networks. Extra Radford et al,

Transfer Learning Enhanced Generative Adversarial Networks

Generative Adversarial Networks (GANs)

3D Generative Adversarial Networks inference

Generative Adversarial Networks - Ian Goodfello · Generative Adversarial Networks 1 ... peaked distributions -> slow mixing • Why poor mixing? ... 2015_icml_gans_wkshp_invited

Heterogeneous Cellular Networks - CATTpanwar/recent_talks/Heterogeneous Cellular... · congestion for voice ... heterogeneous cellular networks ... – Unplanned deployments increase

Adversarial Learning on Heterogeneous Information Networks · Adversarial Learning on Heterogeneous Information Networks Binbin Hu1,2†, Yuan Fang3, Chuan Shi1‡ 1Beijing University

GENERATIVE ADVERSARIAL NETWORKS - Deep Learning

Adversarial Heterogeneous Network Embeddingwith Metapath ...jcst.ict.ac.cn/fileup/1000-9000/PDF/2019-6-4-9427.pdf · porate structural information to embed homogeneous networks. However,

Time-series Generative Adversarial Networks

Deep Video Stabilization Using Adversarial Networks

Perceptual Generative Adversarial Networks for Small ...openaccess.thecvf.com/content...Generative_Adversarial_CVPR_2017… · Perceptual Generative Adversarial Networks for Small

Hypothetical Exploration on Generative Adversarial Networks · Convolutional Neural Networks, Generative Adversarial Net- works, Laplacian Pyramid of Adversarial Networks. I. INTRODUCTION

Correction of Adversarial Errors in Networks

Collaborative Sampling in Generative Adversarial … › pdf › 1902.00813.pdfGenerative Adversarial Networks Generative Adversarial Networks (GANs) (Goodfellow et al. 2014) consist

Domain Adaptation with Adversarial Neural Networks and ... · DAP: Domain Adaptation with Adversarial Neural Networks and Auto-encoders Domain Adaptation with Adversarial Neural Networks