DSGAN: Generative Adversarial Training for Distant Supervision Relation Extraction

Qin, Pengda; Xu, Weiran; Wang, William Yang

doi:10.18653/v1/p18-1046

Cited by 152 publications

(101 citation statements)

References 21 publications

Supporting

Mentioning

101

Contrasting

Order By: Relevance

“…Adversarial training has also been adopted for information extraction (Wu et al, 2017;Hong et al, 2018;Qin et al, 2018;. These adversarial information extraction methods either generate adversarial instances by adding simple noise perturbation to embeddings (Wu et al, 2017;Hong et al, 2018), or mainly adopt models to denoise data and neglect to discover more training instances from raw data (Qin et al, 2018;. Compared with these methods, our adversarial method samples adversarial examples from the real-world data rather than generating pseudo noisy perturbations.…”

Section: Related Workmentioning

confidence: 99%

Adversarial Training for Weakly Supervised Event Detection

Wang

Han

Liu

et al. 2019

Proceedings of the 2019 Conference of the North

118

View full text Add to dashboard Cite

Modern weakly supervised methods for event detection (ED) avoid time-consuming human annotation and achieve promising results by learning from auto-labeled data. However, these methods typically rely on sophisticated pre-defined rules as well as existing instances in knowledge bases for automatic annotation and thus suffer from low coverage, topic bias, and data noise. To address these issues, we build a large event-related candidate set with good coverage and then apply an adversarial training mechanism to iteratively identify those informative instances from the candidate set and filter out those noisy ones. The experiments on two real-world datasets show that our candidate selection and adversarial training can cooperate together to obtain more diverse and accurate training data for ED, and significantly outperform the state-of-theart methods in various weakly supervised scenarios. The datasets and source code can be obtained from https://github.com/ thunlp/Adv-ED.

show abstract

Section: Related Workmentioning

confidence: 99%

Adversarial Training for Weakly Supervised Event Detection

Wang

Han

Liu

et al. 2019

Proceedings of the 2019 Conference of the North

118

View full text Add to dashboard Cite

show abstract

“…[30] proposes to model the noise caused by wrong labelling problem and show that dynamic transition matrix can effectively characterize the noises. [31,32] propose to use adversarial learning [33] to solve the wrong labelling problem. Instead, [34,35] adopt reinforcement learning to learn to select high-quality data for training.…”

Section: Recent Years Deep Learning Has Achieved Remarkable Success mentioning

confidence: 99%

Deep ranking based cost-sensitive multi-label learning for distant supervision relation extraction

Hai

Luo

2020

Information Processing & Management

View full text Add to dashboard Cite

Knowledge base provides a potential way to improve the intelligence of information retrieval (IR) systems, for that knowledge base has numerous relations between entities which can help the IR systems to conduct inference from one entity to another entity.Relation extraction is one of the fundamental techniques to construct a knowledge base.Distant supervision is a semi-supervised learning method for relation extraction which learns with labeled and unlabeled data. However, this approach suffers the problem of relation overlapping in which one entity tuple may have multiple relation facts. We believe that relation types can have latent connections, which we call class ties, and can be exploited to enhance relation extraction. However, this property between relation classes has not been fully explored before. In this paper, to exploit class ties between relations to improve relation extraction, we propose a general ranking based multi-label learning framework combined with convolutional neural networks, in which ranking based loss functions with regularization technique are introduced to learn the latent connections between relations. Furthermore, to deal with the problem of class imbalance in distant supervision relation extraction, we further adopt cost-sensitive learning to rescale the costs from the positive and negative labels. Extensive experiments on a widely used dataset show the effectiveness of our model to exploit class ties and to relieve class imbalance problem. Relation extraction (RE) aims to classify the relations (or called relation facts) between two given named entities from natural-language text. Fig. 1 shows two sentences with the same entity tuple but two different relation facts. RE is to accurately extract the corresponding relation facts (place of birth, place lived) for the entity tuple (Patsy Ramsey, Atlanta) based on the contexts of sentences. Supervised-learning methods require numerous labeled data to work well. With the rapid growth of volume of relation types, traditional methods can not keep up with the step for the limitation of labeled data. In order to narrow down the gap of data sparsity, [1] proposes distant supervision (DS) for relation extraction, which automatically generates training data by aligning a knowledge facts database (ie. Freebase [2]) to texts. For a fact (e.g. entity tuple with a relation type) from the knowledge base, the sentences containing the entity tuple in the fact are regarded as the training data.Class ties mean the connections (relatedness) between relations types for relation extraction. In general, we conclude that class ties can have two categories: weak class ties and strong class ties. Weak class ties mainly involve the co-occurrence of relations such as place of birth and place lived, CEO of and founder of. Besides, strong class ties mean that relations have latent logical entailments. Take the two relations of capital of and city of for example, if one entity tuple has the relation of capital of, it must express the relation fact of city of, be...

show abstract

“…com/tyliupku/soft-label-RE. Table 5: AUC values of previous work and our models, where ATT BL+DSGAN and ATT BL+RL are two models proposed in (Qin et al, 2018a) and (Qin et al, 2018b) respectively, † indicates the baseline result reported in (Qin et al, 2018a,b) and ‡ indicates the baseline result given by our implementation.…”

Section: Pr Curvesmentioning

confidence: 99%

Distant Supervision Relation Extraction with Intra-Bag and Inter-Bag Attentions

Ye¹,

Ling²

2019

Proceedings of the 2019 Conference of the North

105

View full text Add to dashboard Cite

This paper presents a neural relation extraction method to deal with the noisy training data generated by distant supervision. Previous studies mainly focus on sentence-level de-noising by designing neural networks with intra-bag attentions. In this paper, both intrabag and inter-bag attentions are considered in order to deal with the noise at sentence-level and bag-level respectively. First, relationaware bag representations are calculated by weighting sentence embeddings using intrabag attentions. Here, each possible relation is utilized as the query for attention calculation instead of only using the target relation in conventional methods. Furthermore, the representation of a group of bags in the training set which share the same relation label is calculated by weighting bag representations using a similarity-based inter-bag attention module. Finally, a bag group is utilized as a training sample when building our relation extractor. Experimental results on the New York Times dataset demonstrate the effectiveness of our proposed intra-bag and inter-bag attention modules. Our method also achieves better relation extraction accuracy than state-of-the-art methods on this dataset 1 .

show abstract

DSGAN: Generative Adversarial Training for Distant Supervision Relation Extraction

Cited by 152 publications

References 21 publications

Adversarial Training for Weakly Supervised Event Detection

Adversarial Training for Weakly Supervised Event Detection

Deep ranking based cost-sensitive multi-label learning for distant supervision relation extraction

Distant Supervision Relation Extraction with Intra-Bag and Inter-Bag Attentions

Contact Info

Product

Resources

About