Integrating Specialized Classifiers Based on Continuous Time Markov Chain
Refereed conference paper presented and published in conference proceedings

香港中文大學研究人員
替代計量分析
.

其它資訊
摘要Specialized classifiers, namely those dedicated to a subset of classes, are often adopted in real-world recognition systems. However, integrating such classifiers is nontrivial. Existing methods, e.g. weighted average, usually depend on an implicit assumption that all constituents of an ensemble cover the same set of classes. Such methods can produce misleading predictions when used to combine specialized classifiers. This work explores a novel approach. Instead of combining predictions from individual classifiers directly, it first decomposes the predictions into sets of pairwise preferences, treating them as transition channels between classes, and thereon constructs a continuous-time Markov chain. This way allows us to form a coherent picture over all specialized predictions. On large public datasets, the proposed method obtains considerable improvement compared to mainstream ensemble methods, especially when the classifier coverage is highly unbalanced.
著者Zhizhong Li, Dahua Lin
會議名稱International Joint Conference on Artificial Intelligence (IJCAI)
會議開始日19.08.2017
會議完結日25.08.2017
會議地點Melbourne
會議國家/地區澳大利亞
會議論文集題名Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence
出版年份2017
月份8
頁次2244 - 2251
語言美式英語

上次更新時間 2021-16-01 於 00:56