Progressive Memory Banks for Incremental Domain Adaptation

Abstract

This paper addresses the problem of incremental domain adaptation (IDA) in natural language processing (NLP). We assume each domain comes one after another, and that we could only access data in the current domain. The goal of IDA is to build a unified model performing well on all the domains that we have encountered. We adopt the recurrent neural network (RNN) widely used in NLP, but augment it with a directly parameterized memory bank, which is retrieved by an attention mechanism at each step of RNN transition. The memory bank provides a natural way of IDA: when adapting our model to a new domain, we progressively add new slots to the memory bank, which increases the number of parameters, and thus the model capacity. We learn the new memory slots and fine-tune existing parameters by back-propagation. Experimental results show that our approach achieves significantly better performance than fine-tuning alone. Compared with expanding hidden states, our approach is more robust for old domains, shown by both empirical and theoretical results. Our model also outperforms previous work of IDA including elastic weight consolidation and progressive neural networks in the experiments.

arXiv commentICLR 2020

References in corpus (14)

• Sequence to Sequence Learning with Neural Networks
• Overcoming catastrophic forgetting in neural networks
• Continual Learning Through Synaptic Intelligence
• End-To-End Memory Networks
• Efficient Lifelong Learning with A-GEM
• Chameleons in imagined conversations: A new approach to understanding coordination of linguistic style in dialogs
• Overcoming Catastrophic Forgetting by Incremental Moment Matching
• Reinforced Continual Learning
• Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization
• Memory Replay GANs: learning to generate images from new categories without forgetting
• On Fast Dropout and its Applicability to Recurrent Networks
• Scalable and Order-robust Continual Learning with Additive Parameter Decomposition
• Overcoming catastrophic forgetting problem by weight consolidation and long-term memory
• Incremental Learning with Maximum Entropy Regularization: Rethinking Forgetting and Intransigence

Cited by in corpus (3)

• Continual Learning for Recurrent Neural Networks: an Empirical Evaluation
• Continual Learning with Gated Incremental Memories for sequential data processing
• Continual Learning in Recurrent Neural Networks