Continuous Learning for Large-scale Personalized Domain Classification
This addresses the problem of scaling domain classification for industry IPDAs with rapidly growing third-party domains, though it appears incremental as it builds on existing continual learning methods.
The paper tackles the challenge of continuously accommodating new domains in intelligent personal digital assistants by proposing CoNDA, a neural network approach for domain classification that supports incremental learning, achieving high accuracy and outperforming baselines by a large margin.
Domain classification is the task of mapping spoken language utterances to one of the natural language understanding domains in intelligent personal digital assistants (IPDAs). This is a major component in mainstream IPDAs in industry. Apart from official domains, thousands of third-party domains are also created by external developers to enhance the capability of IPDAs. As more domains are developed rapidly, the question of how to continuously accommodate the new domains still remains challenging. Moreover, existing continual learning approaches do not address the problem of incorporating personalized information dynamically for better domain classification. In this paper, we propose CoNDA, a neural network based approach for domain classification that supports incremental learning of new classes. Empirical evaluation shows that CoNDA achieves high accuracy and outperforms baselines by a large margin on both incrementally added new domains and existing domains.