Learning Data Manipulation for Augmentation and Weighting

Abstract

Manipulating data, such as weighting data examples or augmenting with new instances, has been increasingly used to improve model training. Previous work has studied various rule- or learning-based approaches designed for specific types of data manipulation. In this work, we propose a new method that supports learning different manipulation schemes with the same gradient-based algorithm. Our approach builds upon a recent connection of supervised learning and reinforcement learning (RL), and adapts an off-the-shelf reward learning algorithm from RL for joint data manipulation learning and model training. Different parameterization of the "data reward" function instantiates different manipulation schemes. We showcase data augmentation that learns a text transformation network, and data weighting that dynamically adapts the data sample importance. Experiments show the resulting algorithms significantly improve the image and text classification performance in low data regime and class-imbalance problems.

NeurIPS 2019

Cited by in corpus (28)

• A Survey on Data Augmentation for Text Classification
• Data Augmentation Approaches in Natural Language Processing: A Survey
• Data Augmentation in Natural Language Processing: A Novel Text Generation Approach for Long and Short Text Classifiers
• Data Augmentation using Pre-trained Transformer Models
• Model Patching: Closing the Subgroup Performance Gap with Data Augmentation
• CoDA: Contrast-enhanced and Diversity-promoting Data Augmentation for Natural Language Understanding
• On the Generalization Effects of Linear Transformations in Data Augmentation
• Improving GAN Training with Probability Ratio Clipping and Sample Reweighting
• Learning Fast Sample Re-weighting Without Reward Data
• Exploiting All Samples in Low-Resource Sentence Classification: Early Stopping and Initialization Parameters
• DoubleEnsemble: A New Ensemble Method Based on Sample Reweighting and Feature Selection for Financial Data Analysis
• Bowtie Networks: Generative Modeling for Joint Few-Shot Recognition and Novel-View Synthesis
• Generalized Data Weighting via Class-level Gradient Manipulation
• Teaching with Commentaries
• AutoQA: From Databases To QA Semantic Parsers With Only Synthetic Training Data
• Data Manipulation: Towards Effective Instance Learning for Neural Dialogue Generation via Learning to Augment and Reweight
• IOT: Instance-wise Layer Reordering for Transformer Structures
• BET: A Backtranslation Approach for Easy Data Augmentation in Transformer-based Paraphrase Identification Context
• AUGNLG: Few-shot Natural Language Generation using Self-trained Data Augmentation
• Summarizing Text on Any Aspects: A Knowledge-Informed Weakly-Supervised Approach
• Learning to Augment for Data-Scarce Domain BERT Knowledge Distillation
• Data Augmentation for Text Generation Without Any Augmented Data
• Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision
• Stronger Transformers for Neural Multi-Hop Question Generation
• Meta-Learning to Improve Pre-Training
• AEDA: An Easier Data Augmentation Technique for Text Classification
• Text AutoAugment: Learning Compositional Augmentation Policy for Text Classification
• A Diversity-Enhanced and Constraints-Relaxed Augmentation for Low-Resource Classification