Learn-Prune-Share for Lifelong Learning
arXiv:2012.06956
Abstract
In lifelong learning, we wish to maintain and update a model (e.g., a neural network classifier) in the presence of new classification tasks that arrive sequentially. In this paper, we propose a learn-prune-share (LPS) algorithm which addresses the challenges of catastrophic forgetting, parsimony, and knowledge reuse simultaneously. LPS splits the network into task-specific partitions via an ADMM-based pruning strategy. This leads to no forgetting, while maintaining parsimony. Moreover, LPS integrates a novel selective knowledge sharing scheme into this ADMM optimization framework. This enables adaptive knowledge sharing in an end-to-end fashion. Comprehensive experimental results on two lifelong learning benchmark datasets and a challenging real-world radio frequency fingerprinting dataset are provided to demonstrate the effectiveness of our approach. Our experiments show that LPS consistently outperforms multiple state-of-the-art competitors.
arXiv commentAccepted to the IEEE International Conference on Data Mining 2020 (ICDM'20)
References in corpus (10)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Overcoming catastrophic forgetting in neural networks
- Continual Learning Through Synaptic Intelligence
- Rethinking the Value of Network Pruning
- An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks
- A Systematic DNN Weight Pruning Framework using Alternating Direction Method of Multipliers
- Generative replay with feedback connections as a general strategy for continual learning
- Continual Learning via Neural Pruning
- Neurogenesis Deep Learning
- Progressive Weight Pruning of Deep Neural Networks using ADMM