Adversarial Multi-task Learning for Text Classification

Abstract

Neural network models have shown their promising opportunities for multi-task learning, which focus on learning the shared layers to extract the common and task-invariant features. However, in most existing approaches, the extracted shared features are prone to be contaminated by task-specific features or the noise brought by other tasks. In this paper, we propose an adversarial multi-task learning framework, alleviating the shared and private latent feature spaces from interfering with each other. We conduct extensive experiments on 16 different text classification tasks, which demonstrates the benefits of our approach. Besides, we show that the shared knowledge learned by our proposed model can be regarded as off-the-shelf knowledge and easily transferred to new tasks. The datasets of all 16 tasks are publicly available at \url{http://nlp.fudan.edu.cn/data/}

Accepted by ACL2017

References in corpus (8)

• Sequence to Sequence Learning with Neural Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales
• Domain Separation Networks
• A Convolutional Neural Network for Modelling Sentences
• Unsupervised Cross-Domain Image Generation
• Domain-Adversarial Neural Networks
• Image-Text Multi-Modal Representation Learning by Adversarial Backpropagation

Cited by in corpus (21)

• Multi-Task Learning with Deep Neural Networks: A Survey
• Deep Learning Based Text Classification: A Comprehensive Review
• HireNet: a Hierarchical Attention Model for the Automatic Analysis of Asynchronous Video Job Interviews
• Neural Text Generation: Past, Present and Beyond
• Adversarial multi-task underwater acoustic target recognition: towards robustness against various influential factors
• Is Discriminator a Good Feature Extractor?
• Face Completion with Semantic Knowledge and Collaborative Adversarial Learning
• Joint Multi-Domain Learning for Automatic Short Answer Grading
• Estimation of Inter-Sentiment Correlations Employing Deep Neural Network Models
• Modelling Domain Relationships for Transfer Learning on Retrieval-based Question Answering Systems in E-commerce
• Style Obfuscation by Invariance
• Domain-Invariant Feature Distillation for Cross-Domain Sentiment Classification
• Adversarial Training Based Multi-Source Unsupervised Domain Adaptation for Sentiment Analysis
• Multi-task Learning by Leveraging the Semantic Information
• Unsupervised Sentiment Analysis by Transferring Multi-source Knowledge
• A Label Proportions Estimation Technique for Adversarial Domain Adaptation in Text Classification
• Empirical Evaluation of Multi-task Learning in Deep Neural Networks for Natural Language Processing
• Multiple Face Analyses through Adversarial Learning
• Multi-Task Generative Adversarial Nets with Shared Memory for Cross-Domain Coordination Control
• Emotion Correlation Mining Through Deep Learning Models on Natural Language Text
• A Short Review on Data Modelling for Vector Fields