具有漸進(jìn)局部學(xué)習(xí)特性的多色Voronoi分類器設(shè)計(jì)
Design of Multicolor Voronoi Classifier with Gradually Local Learning Ablility
-
摘要: 本文提出了一種多色Voronoi分類器MCVC。MCVC在學(xué)習(xí)樣本上有好的邊界推廣性隨著樣本數(shù)量的增加MCVC的分類面可以逼近任意的分類函數(shù)。MCVC具有好的局部特性，對新加樣本的訓(xùn)練只影響其周圍的局部性態(tài)，不會(huì)對全局產(chǎn)生大的影響，可以克服神經(jīng)網(wǎng)絡(luò)方法對樣本的過學(xué)習(xí)問題。實(shí)驗(yàn)表明MCVC對于線性和非線性分類問題都具有最優(yōu)分類面。
-
關(guān)鍵詞:
- Voronoi圖; 多色; 分類器設(shè)計(jì)
Abstract: A novel MultiColor Voronoi Classifier (MCVC) is proposed, which can be ap-plied to linear and nonlinear classification problems. MCVC has sound ability to expend classification plane between samples. With increment of samples, it can be shown that the classification plane of MCVC can close to any classification function. MCVC has very good local ability too. When new learning sample is added, only local classification planed is modified and the whole classification characteristics are not modified greatly. So MCVC can solve the overfitting problem of neural network. Experiments show that MCVC is feasible to linear classification and nonlinear classification problems. -
蔡元龍.模式識別.西安:西安電子科技大學(xué)出版社,1986年,第3章:67-101.[2]邊肇祺,張學(xué)工.模式識別.北京:清華大學(xué)出版社,2000年,第7章:161-173;第13章:284-303.[3]Vapnik V N,張學(xué)工譯.統(tǒng)計(jì)學(xué)習(xí)理論的本質(zhì).北京:清華大學(xué)出版社,2000年,第0章:1-10.[4]Vapnik V N. An overview of statistical learning theory[J].IEEE Trans. Neural Networks.1999,10(5):988-999[5]周培德.計(jì)算幾何-算法分析與設(shè)計(jì).北京:清華大學(xué)出版社,2000年,第4章,88-132;第10章: 236-271.[6]Fortune S. A sweepline algorithm for Voronoi diagrams[J].Algorithmica.1987, 2(2):153-174[7]Chen D Z. Efficient geometric algorithm on the EREW PRAM[J].IEEE Trans. Parallel Distrib.Syst.1995, 6(1):41-47[8]Shamos M I, Hoey D. Closest-point problems. Proc. 16tn IEEE Ann. Symp. on the Foundations of Computer Science, CA, USA, 1975: 151-162.[9]Amato N M, Preparata F P. An NC parallel 3D Convex hull algorithm. In: Proc. 19th Annual ACM Symp. Comput. Geom., San Diego, CA, USA, 1993: 289-297.[10]Amato N M, Goodrich M T, Ramos E A. Parallel algorithms for higher-dimensional convex hulls.In: Proc. 35th Annual IEEE Symp. Found. Comput. Sci., Santa Fe, NM, USA, 1994: 386-694.[11]Amato N M, Goodrich M T, Ramos E A. Computing faces in segment and simplex arrangements.In: Proc. 27th Annual ACM Symp. Theory Comput., Las Vegas, Nevada, USA, 1995: 672-682. -
計(jì)量
- 文章訪問數(shù): 2233
- HTML全文瀏覽量: 145
- PDF下載量: 587
- 被引次數(shù): 0
下載:
