Life-long learning based on dynamic combination model
| Autor: | Hong Gao, Jianzhong Li, Hongzhi Wang, Boya Ren |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Incremental decision tree
Computer science Binary decision diagram Data stream mining business.industry Decision tree learning Decision tree 02 engineering and technology Semi-supervised learning Overfitting Machine learning computer.software_genre 020204 information systems Incremental learning 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence Data mining Pruning (decision trees) business computer Software |
| Zdroj: | Applied Soft Computing. 56:398-404 |
| ISSN: | 1568-4946 |
| Popis: | HighlightsA model which can self-adaptively evolve to adjust to the changing data distribution by combining the sub-models after new training.The proposed model can be learned without referring to previous data.Hypothesis are combined nonlinearly using binary decision tree to achieve enhanced performance.The model size is strictly restricted through multiple strategies. In this paper, we propose a novel life-long learning framework, which constantly evolves with changing data distribution, learning new knowledge while retaining some old knowledge. In many practical systems, data in the past is still useful but no longer available. Therefore, a question arises on how to update the model based on both new data and current model. To address this issue, our framework lays its basis on ensemble method with multiple sub-classifiers, independent of base type. When new data is processed, new sub-classifiers are generated accordingly. The classifiers are then dynamically combined using decision tree, together with a novelly proposed pruning method to prevent overfitting and eliminate out-dated models. Guarantees are provided to the combination method. Experiments indicate that the framework achieves good performance when the data changes with time, and has better accuracy compared to existing transfer and incremental learning, and methods in stream data mining. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect