Learning with Noise: Enhance Distantly Supervised Relation Extraction with Dynamic Transition Matrix

Abstract

Distant supervision significantly reduces human efforts in building training data for many classification tasks. While promising, this technique often introduces noise to the generated training data, which can severely affect the model performance. In this paper, we take a deep look at the application of distant supervision in relation extraction. We show that the dynamic transition matrix can effectively characterize the noise in the training data built by distant supervision. The transition matrix can be effectively trained using a novel curriculum learning based method without any direct supervision about the noise. We thoroughly evaluate our approach under a wide range of extraction scenarios. Experimental results show that our approach consistently improves the extraction results and outperforms the state-of-the-art in various evaluation scenarios.

arXiv comment10 pages, accepted by ACL 2017

References in corpus (4)

• Distributed Representations of Words and Phrases and their Compositionality
• Training Deep Neural Networks on Noisy Labels with Bootstrapping
• Training Convolutional Networks with Noisy Labels
• Webly Supervised Learning of Convolutional Networks

Cited by in corpus (21)

• Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages
• FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation
• WRENCH: A Comprehensive Benchmark for Weak Supervision
• A Question-answering Based Framework for Relation Extraction Validation
• Improving Distantly Supervised Relation Extraction with Neural Noise Converter and Conditional Optimal Selector
• Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approach
• Learning to Denoise Distantly-Labeled Data for Entity Typing
• DIAG-NRE: A Neural Pattern Diagnosis Framework for Distantly Supervised Neural Relation Extraction
• Cross-relation Cross-bag Attention for Distantly-supervised Relation Extraction
• DSReg: Using Distant Supervision as a Regularizer
• O2D2: Out-Of-Distribution Detector to Capture Undecidable Trials in Authorship Verification
• Feature-Dependent Confusion Matrices for Low-Resource NER Labeling with Noisy Labels
• A Practical Framework for Relation Extraction with Noisy Labels Based on Doubly Transitional Loss
• Decoupled Gradient Harmonized Detector for Partial Annotation: Application to Signet Ring Cell Detection
• Representation Learning from Limited Educational Data with Crowdsourced Labels
• Uncover the Ground-Truth Relations in Distant Supervision: A Neural Expectation-Maximization Framework
• Deep Ranking Based Cost-sensitive Multi-label Learning for Distant Supervision Relation Extraction
• Distantly Supervised Relation Extraction in Federated Settings
• Abstractified Multi-instance Learning (AMIL) for Biomedical Relation Extraction
• From Bag of Sentences to Document: Distantly Supervised Relation Extraction via Machine Reading Comprehension
• Towards Time-Aware Distant Supervision for Relation Extraction