Compacting, Picking and Growing for Unforgetting Continual Learning

Abstract

Continual lifelong learning is essential to many applications. In this paper, we propose a simple but effective approach to continual deep learning. Our approach leverages the principles of deep model compression, critical weights selection, and progressive networks expansion. By enforcing their integration in an iterative manner, we introduce an incremental learning method that is scalable to the number of sequential tasks in a continual learning process. Our approach is easy to implement and owns several favorable characteristics. First, it can avoid forgetting (i.e., learn new tasks while remembering all previous tasks). Second, it allows model expansion but can maintain the model compactness when handling sequential tasks. Besides, through our compaction and selection/expansion mechanism, we show that the knowledge accumulated through learning previous tasks is helpful to build a better model for the new tasks compared to training the models independently with tasks. Experimental results show that our approach can incrementally learn a deep model tackling multiple tasks without forgetting, while the model compactness is maintained with the performance more satisfiable than individual task training.

To appear in NeurIPS 2019

References in corpus (1)

• Large-scale Classification of Fine-Art Paintings: Learning The Right Metric on The Right Feature

Cited by in corpus (7)

• Continual Learning in Sensor-based Human Activity Recognition: an Empirical Benchmark Analysis
• Generalizing MLPs With Dropouts, Batch Normalization, and Skip Connections
• GROWN: GRow Only When Necessary for Continual Learning
• KSM: Fast Multiple Task Adaption via Kernel-wise Soft Mask Learning
• Compression-aware Continual Learning using Singular Value Decomposition
• Discriminative Distillation to Reduce Class Confusion in Continual Learning
• Continual Learning for Task-oriented Dialogue System with Iterative Network Pruning, Expanding and Masking