Lifelong Intent Detection via Multi-Strategy Rebalancing
This addresses the challenge of continually emerging user intents in online systems, which is a domain-specific problem for intent detection applications, though it is incremental as it builds on existing lifelong learning methods.
The paper tackles the problem of lifelong intent detection, where models must learn new intents over time without forgetting old ones, and proposes a Multi-Strategy Rebalancing method that significantly outperforms previous state-of-the-art methods on benchmarks like ATIS, SNIPS, HWU64, and CLINC150.
Conventional Intent Detection (ID) models are usually trained offline, which relies on a fixed dataset and a predefined set of intent classes. However, in real-world applications, online systems usually involve continually emerging new user intents, which pose a great challenge to the offline training paradigm. Recently, lifelong learning has received increasing attention and is considered to be the most promising solution to this challenge. In this paper, we propose Lifelong Intent Detection (LID), which continually trains an ID model on new data to learn newly emerging intents while avoiding catastrophically forgetting old data. Nevertheless, we find that existing lifelong learning methods usually suffer from a serious imbalance between old and new data in the LID task. Therefore, we propose a novel lifelong learning method, Multi-Strategy Rebalancing (MSR), which consists of cosine normalization, hierarchical knowledge distillation, and inter-class margin loss to alleviate the multiple negative effects of the imbalance problem. Experimental results demonstrate the effectiveness of our method, which significantly outperforms previous state-of-the-art lifelong learning methods on the ATIS, SNIPS, HWU64, and CLINC150 benchmarks.