本周论文推荐（12.14-12.20）

导读

下面论文列表为本小组同学推荐和整理，关于论文细节请联系论文作者。内容 涵盖以下研究方向：对抗生成网络、知识图谱补全、推荐系统、文本生成、元学习。

来自：RUC AI Box 实验室

本周论文推荐

 1  

 对抗生成网络 

• Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks. Vineet Kosaraju (Stanford University) · Amir Sadeghian (Stanford University) · Roberto Martín-Martín (Stanford University) · Ian Reid (University of Adelaide) · Hamid Rezatofighi (Stanford University // University of Adelaide) · Silvio Savarese (Stanford University). NeurIPS 2019

  
  

• Twin Auxilary Classifiers GAN. Mingming Gong (University of Melbourne) · Yanwu Xu (University of Pittsburgh) · Chunyuan Li (Microsoft Research) · Kun Zhang (CMU) · Kayhan Batmanghelich (University of Pittsburgh). NeurIPS 2019

• Multi-marginal Wasserstein GAN. Jiezhang Cao (South China University of Technology) · Langyuan Mo (South China University of Technology) · Yifan Zhang (South China University of Technology) · Kui Jia (South China University of Technology) · Chunhua Shen (University of Adelaide) · Mingkui Tan (South China University of Technology). NeurIPS 2019

  
  

• Training Language GANs from Scratch. Cyprien de Masson d'Autume (Google DeepMind) · Shakir Mohamed (DeepMind) · Mihaela Rosca (Google DeepMind) · Jack Rae (DeepMind, UCL). NeurIPS 2019

 2  

 知识图谱补全 

• Collaborative Policy Learning for Open Knowledge Graph Reasoning. Cong Fu, Tong Chen, Meng Qu, Woojeong Jin and Xiang Ren. EMNLP 2019

• DIVINE: A Generative Adversarial Imitation Learning Framework for Knowledge Graph Reasoning. Ruiping Li and Xiang Cheng. EMNLP 2019

  
  

• Incorporating Graph Attention Mechanism into Knowledge Graph Reasoning Based on Deep Reinforcement Learning. Xin Lv, Yuxian Gu, Xu Han, Lei Hou, Juanzi Li and Zhiyuan Liu. EMNLP 2019

 3  

 推荐系统

• Reinforcement Knowledge Graph Reasoning for Explainable Recommendation. Yikun Xian, Zuohui Fu, S. Muthukrishnan, Gerard de Melo and Yongfeng Zhang. SIGIR 2019

  
  

• A Model-Based Reinforcement Learning with Adversarial Training for Online Recommendation. Xueying Bai, Jian Guan, Hongning Wang. NeurIPS 2019

  
  

• Text-Based Interactive Recommendation via Constraint-Augmented Reinforcement Learning. Ruiyi Zhang, Tong Yu, Yilin Shen, Hongxia Jin, Changyou Chen. NeurIPS 2019

  
  

 4  

 文本生成

• RelGAN: Relational Generative Adversarial Networks for Text Generation. Weili Nie, Nina Narodytska, Ankit Patel. ICLR 2019
  

• Adversarial Domain Adaptation Using Artificial Titles for Abstractive Title Generation. Francine Chen, Yan-Ying Chen. ACL 2019

• ARAML: A Stable Adversarial Training Framework for Text Generation. Pei Ke, Fei Huang, Minlie Huang, Xiaoyan Zhu. EMNLP 2019

• Enhancing Topic-to-Essay Generation with External Commonsense Knowledge. Yang, Pengcheng and Li, Lei and Luo, Fuli and Liu, Tianyu and Sun, Xu. ACL 2019

• GNNExplainer: Generating Explanations for Graph Neural Networks. Zhitao Ying, Dylan Bourgeois, Jiaxuan You, Marinka Zitnik, Jure Leskovec. NeurIPS 2019

  
  

 5  

 元学习

• Meta-Learning Update Rules for Unsupervised Representation Learning. Luke Metz, Niru Maheswaranathan, Brian Cheung, Jascha Sohl-Dickstein. ICLR 2019

• Adversarial Attacks on Graph Neural Networks via Meta Learning. Daniel Zügner, Stephan Günnemann. ICLR 2019

• Meta-Learning with Latent Embedding Optimization. Andrei A. Rusu, Dushyant Rao, Jakub Sygnowski, Oriol Vinyals, Razvan Pascanu, Simon Osindero, Raia Hadsell. ICLR 2019

  
  

• Unsupervised Learning via Meta-Learning. Kyle Hsu, Sergey Levine, Chelsea Finn. ICLR 2019

  
  

本期编辑：胡晓璇