DSGAN: Generative Adversarial Training for Distant Supervision Relation Extraction

By Pengda Qin, Weiran Xu, William Yang Wang. 2018

159 citations recorded by the source corpus.

Distant supervision can effectively label data for relation extraction, but suffers from the noise labeling problem. Recent works mainly perform soft bag-level noise reduction strategies to find the relatively better samples in a sentence bag, which is suboptimal compared with making a hard decision of false positive samples in sentence level. In this paper, we introduce an adversarial learning framework, which we named DSGAN, to learn a sentence-level true-positive generator. Inspired by Generative Adversarial Networks, we regard the positive samples generated by the generator as the negative samples to train the discriminator. The optimal generator is obtained until the discrimination ability of the discriminator has the greatest decline. We adopt the generator to filter distant supervision training dataset and redistribute the false positive instances into the negative set, in which way to provide a cleaned dataset for relation classification. The experimental results show that the proposed strategy significantly improves the performance of distant supervision relation extraction comparing to state-of-the-art systems.

Subfields: Artificial Intelligence, Natural Language Processing, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Dacheng Tao
2. Philip S. Yu
3. Mohit Bansal
4. Erik Cambria
5. Jianfeng Gao
6. Qun Liu
7. Xin Jiang
8. Hao Peng

In Natural Language Processing

1. Mohit Bansal
2. Adyasha Maharana
3. Erik Cambria
4. Qun Liu
5. Philip S. Yu
6. Zhenhua Ling
7. Hao Peng
8. Zhiquan Ye

In Generative Models

1. Eric I-Chao Chang
2. Dacheng Tao
3. Mohit Bansal
4. Jie Gui
5. Philip S. Yu
6. Jieping Ye
7. Yonggang Wen
8. Qun Liu

Citations in the corpus

• Large Scale Image Completion via Co-Modulated Generative Adversarial Networks (2021, background): “A branch of works aims to enforce the intra-conditioning diversity using VAE-based latent sampling strategies (Zhu et al., 2017b) or imposing distance-based loss terms (Huang et al., 2018; Mao et al., 2019; Qin et al., 2018).”
• GraDA: Graph Generative Data Augmentation for Commonsense Reasoning (2022, methodology): “Most NLP applications use REINFORCE (Sutton et al., 2000) algorithm and its variants (Yu et al., 2017; Cai and Wang, 2018; Qin et al., 2018; Zhang et al., 2018) to circumvent the discrete sampling issue for text-based GANs.”
• Prototypical Representation Learning for Relation Extraction (2021, unknown): “…-10.900574 32.073906 -7.5629783 36.904022 -15.117426 -47.537777 , various denoising methods have been explored to reduce noise effects for distant supervision (Zeng et al., 2015; Lin et al., 2016; Jiang et al., 2016; Han et al., 2018a; Wu et al., 2017; Qin et al., 2018a;b; Feng et al., 2018).”
• GraDA: Graph Generative Data Augmentation for Commonsense Reasoning (2022, methodology): “, 2000) algorithm and its variants (Yu et al., 2017; Cai and Wang, 2018; Qin et al., 2018; Zhang et al., 2018) to circumvent the discrete sampling issue for text-based GANs.”
• Long-tail Relation Extraction via Knowledge Graph Embeddings and Graph Convolution Networks (2019, background): “…further improve the performance, some studies incorporate external information (Zeng et al., 2017; Ji et al., 2017; Han et al., 2018a) and advanced training strategies (Ye et al., 2017; Liu et al., 2017; Huang and Wang, 2017; Feng et al., 2018; Zeng et al., 2018; Wu et al., 2017; Qin et al., 2018).”
• Distant Supervision Relation Extraction with Intra-Bag and Inter-Bag Attentions (2019, methodology): “And also, similar to (Qin et al., 2018b), we only apply inter-bag attentions to positive samples, i.e., the bags whose relation label is not NA ( NoRelation ).”
• Dialog State Tracking with Reinforced Data Augmentation (2019, methodology): “Previous works by (Feng et al., 2018) and (Qin et al., 2018b) design RL algorithm to learn how to fil-ter out noisy ones.”
• A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, methodology): “Inspired by the diachronic word embedding, Goel et al. [182] CNN + max pooling position embedding Multi CNN [137] Multi-window convolution + max pooling position embedding PCNN [138] CNN + piecewise max pooling position embedding MIMLCNN [139] CNN + piecewise and cross-sentence max pooling position ”
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications (2020, background): “DSGAN [388] is proposed for distant supervision relation extraction.”
• Open Event Extraction from Online Text using a Generative Adversarial Network (2019, background): “(2018) improved the performance of semi-supervised text classification using adversarial training, (Zeng et al., 2018; Qin et al., 2018) designed GAN-based models for distance supervision relation extraction.”
• A Comprehensive Survey on Automatic Knowledge Graph Construction (2023, methodology): “PCNN+pooling positionembeddings TM[190] PCNN+Transitionmatrix+Bagembedding positionembeddings HAtt[189] CNN/PCNN+hierarchicalattention positionembeddings,relationhierarchy Enhancedlearningstrategy-based RL-based[192] RL-baseddataredistributor+CNN/PCNN+Result-drivenreward positionembeddings DSGAN[193”