Continual Lifelong Learning in Natural Language Processing: A Survey

Abstract

Continual learning (CL) aims to enable information systems to learn from a continuous data stream across time. However, it is difficult for existing deep learning architectures to learn a new task without largely forgetting previously acquired knowledge. Furthermore, CL is particularly challenging for language learning, as natural language is ambiguous: it is discrete, compositional, and its meaning is context-dependent. In this work, we look at the problem of CL through the lens of various NLP tasks. Our survey discusses major challenges in CL and current methods applied in neural network models. We also provide a critical review of the existing CL evaluation methods and datasets in NLP. Finally, we present our outlook on future research directions.

COLING 2020

Cited by in corpus (11)

• Class-Incremental Learning: A Survey
• AI in Human-computer Gaming: Techniques, Challenges and Opportunities
• Inherit with Distillation and Evolve with Contrast: Exploring Class Incremental Semantic Segmentation Without Exemplar Memory
• Using Adapters to Overcome Catastrophic Forgetting in End-to-End Automatic Speech Recognition
• Learning at a Glance: Towards Interpretable Data-limited Continual Semantic Segmentation via Semantic-Invariance Modelling
• Unified Question Generation with Continual Lifelong Learning
• Weight Averaging: A Simple Yet Effective Method to Overcome Catastrophic Forgetting in Automatic Speech Recognition
• DRILL: Dynamic Representations for Imbalanced Lifelong Learning
• VIPeR: Visual Incremental Place Recognition with Adaptive Mining and Continual Learning
• Analyzing Multi-Task Learning for Abstractive Text Summarization
• Inverse-Hessian Regularization for Continual Learning in ASR