Accepted Paper: Handling Concept Drift via Model Reuse
Back to list of accepted papers
- Session 6: Supervised Learning -- Day 3 (Nov.19), poster session: 11:30-14:00, talks: 14:10-15:25 (5th floor Hall 2)
- Poster number: Tue31
Authors
Peng Zhao (Nanjing University); Le-Wen Cai (Nanjing University); Zhi-Hua Zhou (Nanjing university)
Abstract
In many real-world applications, data are often collected in the form of a stream, and thus the distribution usually changes in nature, which is referred to as concept drift in the literature. We propose a novel and effective approach to handle concept drift via model reuse, that is, reusing models trained on previous data to tackle the changes. Each model is associated with a weight representing its reusability towards current data, and the weight is adaptively adjusted according to the performance of the model. We provide both generalization and regret analysis. Experimental results also validate the superiority of our approach on both synthetic and real-world datasets.