Enhanced piecewise regression based on deterministic annealing  

作　　者：ZHANG JiangShe YANG YuQian CHEN XiaoWen ZHOU ChengHu 

机构地区：[1]Institute for Information and System Sciences, Faculty of Sciences, Xi'an Jiaotong University, Xi'an 710049, China [2]State Key Laboratory of Resources & Environmental Information System, Institute of Geography, Chinese Academy of Sciences, Beijing 100101, China

出　　处：《Science in China(Series F)》2008年第8期1025-1038,共14页中国科学（F辑英文版）

基　　金：the National Natural Science Foundation of China(Grant Nos.60675013 and 4022500);the National Basic Research Program of China(973 Program)(Grant No.2007CB311002)

摘　　要：Regression is one of the important problems in statistical learning theory. This paper proves the global convergence of the piecewise regression algorithm based on deterministic annealing and continuity of global minimum of free energy w.r.t temperature, and derives a new simplified formula to compute the initial critical temperature. A new enhanced plecewise regression algorithm by using ＂migration of prototypes＂ is proposed to eliminate ＂empty cell＂ in the annealing process. Numerical experiments on several benchmark datasets show that the new algorithm can remove redundancy and improve generalization of the piecewise regression model.Regression is one of the important problems in statistical learning theory. This paper proves the global convergence of the piecewise regression algorithm based on deterministic annealing and continuity of global minimum of free energy w.r.t temperature, and derives a new simplified formula to compute the initial critical temperature. A new enhanced plecewise regression algorithm by using ＂migration of prototypes＂ is proposed to eliminate ＂empty cell＂ in the annealing process. Numerical experiments on several benchmark datasets show that the new algorithm can remove redundancy and improve generalization of the piecewise regression model.

关 键 词：statistical regression piecewise regression deterministic annealing GENERALIZATION 

分 类 号：I17[文学—世界文学]