Update Frequently, Update Fast: Retraining Semantic Parsing Systems in a Fraction of Time
arXiv:2010.07865
Abstract
Currently used semantic parsing systems deployed in voice assistants can require weeks to train. Datasets for these models often receive small and frequent updates, data patches. Each patch requires training a new model. To reduce training time, one can fine-tune the previously trained model on each patch, but naive fine-tuning exhibits catastrophic forgetting - degradation of the model performance on the data not represented in the data patch. In this work, we propose a simple method that alleviates catastrophic forgetting and show that it is possible to match the performance of a model trained from scratch in less than 10% of a time via fine-tuning. The key to achieving this is supersampling and EWC regularization. We demonstrate the effectiveness of our method on multiple splits of the Facebook TOP and SNIPS datasets.
References in corpus (7)
- Overcoming catastrophic forgetting in neural networks
- Scaling Laws for Neural Language Models
- Continual Learning Through Synaptic Intelligence
- Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces
- LAMOL: LAnguage MOdeling for Lifelong Language Learning
- Continual Learning with Gated Incremental Memories for sequential data processing
- Toward Continual Learning for Conversational Agents