Gradient Imitation Reinforcement Learning for Low Resource Relation Extraction

by   Xuming Hu, et al.
Tsinghua University
University of Illinois at Chicago

Low-resource Relation Extraction (LRE) aims to extract relation facts from limited labeled corpora when human annotation is scarce. Existing works either utilize self-training scheme to generate pseudo labels that will cause the gradual drift problem, or leverage meta-learning scheme which does not solicit feedback explicitly. To alleviate selection bias due to the lack of feedback loops in existing LRE learning paradigms, we developed a Gradient Imitation Reinforcement Learning method to encourage pseudo label data to imitate the gradient descent direction on labeled data and bootstrap its optimization capability through trial and error. We also propose a framework called GradLRE, which handles two major scenarios in low-resource relation extraction. Besides the scenario where unlabeled data is sufficient, GradLRE handles the situation where no unlabeled data is available, by exploiting a contextualized augmentation method to generate data. Experimental results on two public datasets demonstrate the effectiveness of GradLRE on low resource relation extraction when comparing with baselines.


page 1

page 2

page 3

page 4


Gradient Imitation Reinforcement Learning for General Low-Resource Information Extraction

Information Extraction (IE) aims to extract structured information from ...

Neural Rule Grounding for Low-Resource Relation Extraction

While deep neural models have gained successes on information extraction...

Towards Realistic Low-resource Relation Extraction: A Benchmark with Empirical Baseline Study

This paper presents an empirical study to build relation extraction syst...

Continual Contrastive Finetuning Improves Low-Resource Relation Extraction

Relation extraction (RE), which has relied on structurally annotated cor...

A Review on Semi-Supervised Relation Extraction

Relation extraction (RE) plays an important role in extracting knowledge...

Relation Adversarial Network for Low Resource KnowledgeGraph Completion

Knowledge Graph Completion (KGC) has been proposed to improve Knowledge ...

Knowledge Extraction in Low-Resource Scenarios: Survey and Perspective

Knowledge Extraction (KE) which aims to extract structural information f...

Please sign up or login with your details

Forgot password? Click here to reset