"산술 기하 평균을 이용한 원주율의 계산"의 두 판 사이의 차이

수학노트
둘러보기로 가기 검색하러 가기
2번째 줄: 2번째 줄:
  
 
* AGM(arithmetic-geometric mean)을 활용하여 파이값을 빠르게 계산할 수 알고리즘
 
* AGM(arithmetic-geometric mean)을 활용하여 파이값을 빠르게 계산할 수 알고리즘
 
 
 
  
 
 
 
 
138번째 줄: 136번째 줄:
 
 
 
 
  
<h5>관련된 다른 주제들<sup style="">[[#toc 8|#]]</sup></h5>
+
<h5>관련된 다른 주제들</h5>
  
 
* [[타원적분|타원적분, 타원함수, 타원곡선]]<br>
 
* [[타원적분|타원적분, 타원함수, 타원곡선]]<br>
 
** [[타원적분(통합됨)|타원적분]]
 
** [[타원적분(통합됨)|타원적분]]
 
** [[렘니스케이트(lemniscate) 곡선의 길이와 타원적분|lemniscate 적분]]
 
** [[렘니스케이트(lemniscate) 곡선의 길이와 타원적분|lemniscate 적분]]
 +
* [[자코비 세타함수]]
 +
* [[#]]
 +
 +
 
  
 
 
 
 
  
<h5>표준적인 도서 및 추천도서<sup style="">[[#toc 9|#]]</sup></h5>
+
<h5>표준적인 도서 및 추천도서</h5>
  
 
* [http://www.amazon.com/PI-AGM-Analytic-Computational-Complexity/dp/047131515X Pi and the AGM]<br>
 
* [http://www.amazon.com/PI-AGM-Analytic-Computational-Complexity/dp/047131515X Pi and the AGM]<br>
155번째 줄: 157번째 줄:
 
 
 
 
  
<h5>위키링크<sup style="">[[#toc 10|#]]</sup></h5>
+
<h5>위키링크</h5>
  
 
* http://en.wikipedia.org/wiki/Arithmetic-geometric_mean
 
* http://en.wikipedia.org/wiki/Arithmetic-geometric_mean
162번째 줄: 164번째 줄:
 
 
 
 
  
<h5>참고할만한 자료<sup style="">[[#toc 11|#]]</sup></h5>
+
 
 +
 
 +
<h5>관련논문</h5>
  
*  Computation of pi Using Arithmetic-Geometric Mean ([[1939326/attachments/1341646|pdf]])<br>
 
** E. Salamin
 
** Mathematics of Computation 30(1976) 565-570
 
 
* [http://wwwmaths.anu.edu.au/%7Ebrent/pub/pub028.html Multiple-precision zero-finding methods and the complexity of elementary function evaluation]<br>
 
* [http://wwwmaths.anu.edu.au/%7Ebrent/pub/pub028.html Multiple-precision zero-finding methods and the complexity of elementary function evaluation]<br>
** R. P. Brent
+
** R. P. Brent, Analytic Computational Complexity (edited by J. F. Traub), Academic Press, New York, 1975, 151–176
** Analytic Computational Complexity (edited by J. F. Traub), Academic Press, New York, 1975, 151–176
 
 
*  The arithmetic-geometric mean of Gauss ([[1939326/attachments/1144114|pdf]])<br>
 
*  The arithmetic-geometric mean of Gauss ([[1939326/attachments/1144114|pdf]])<br>
** D.A. Cox
+
** D.A. Cox, Enseignement Math. 30 (1984) 275-330
** UEnseignement Math. 30 (1984) 275-330
 
 
*  Gauss and the arithmetic-geometric mean<br>
 
*  Gauss and the arithmetic-geometric mean<br>
** D.A. Cox
+
** D.A. Cox, Notices Amer. Math. Soc. 32(2) (1985) 147-151
** Notices Amer. Math. Soc. 32(2) (1985) 147-151
 
  
 
* [http://www.jstor.org/stable/2323302 Gauss, Landen, Ramanujan, the Arithmetic-Geometric Mean, Ellipses, π, and the Ladies Diary]<br>
 
* [http://www.jstor.org/stable/2323302 Gauss, Landen, Ramanujan, the Arithmetic-Geometric Mean, Ellipses, π, and the Ladies Diary]<br>
181번째 줄: 179번째 줄:
 
** <cite>The American Mathematical Monthly</cite>, Vol. 95, No. 7 (Aug. - Sep., 1988), pp. 585-608
 
** <cite>The American Mathematical Monthly</cite>, Vol. 95, No. 7 (Aug. - Sep., 1988), pp. 585-608
 
* [http://www.jstor.org/stable/2325206 Ramanujan, Modular Equations, and Approximations to Pi or How to Compute One Billion Digits of Pi]<br>
 
* [http://www.jstor.org/stable/2325206 Ramanujan, Modular Equations, and Approximations to Pi or How to Compute One Billion Digits of Pi]<br>
** J. M. Borwein, P. B. Borwein and D. H. Bailey
+
** J. M. Borwein, P. B. Borwein and D. H. Bailey<cite>The American Mathematical Monthly</cite>, Vol. 96, No. 3 (Mar., 1989), pp. 201-219
** <cite>The American Mathematical Monthly</cite>, Vol. 96, No. 3 (Mar., 1989), pp. 201-219
 
 
* [http://www.jstor.org/stable/3619132 Recent Calculations of π: The Gauss-Salamin Algorithm]<br>
 
* [http://www.jstor.org/stable/3619132 Recent Calculations of π: The Gauss-Salamin Algorithm]<br>
** Nick Lord
+
** Nick Lord<cite>The Mathematical Gazette</cite>, Vol. 76, No. 476 (Jul., 1992), pp. 231-242
** <cite>The Mathematical Gazette</cite>, Vol. 76, No. 476 (Jul., 1992), pp. 231-242
 
 
* [http://www.jstor.org/stable/2690037 The Ubiquitous π]<br>
 
* [http://www.jstor.org/stable/2690037 The Ubiquitous π]<br>
** Dario Castellanos
+
** Dario Castellanos<cite>Mathematics Magazine</cite>, Vol. 61, No. 2 (Apr., 1988), pp. 67-98
** <cite>Mathematics Magazine</cite>, Vol. 61, No. 2 (Apr., 1988), pp. 67-98
 
 
* [http://www.jstor.org/stable/2031275 The Arithmetic-Geometric Mean and Fast Computation of Elementary Functions]<br>
 
* [http://www.jstor.org/stable/2031275 The Arithmetic-Geometric Mean and Fast Computation of Elementary Functions]<br>
** J. M. Borwein and P. B. Borwein
+
** J. M. Borwein and P. B. Borwein<cite>[http://www.jstor.org/action/showPublication?journalCode=siamreview SIAM Review]</cite>, Vol. 26, No. 3 (Jul., 1984), pp. 351-366
** <cite>[http://www.jstor.org/action/showPublication?journalCode=siamreview SIAM Review]</cite>, Vol. 26, No. 3 (Jul., 1984), pp. 351-366
+
*  Computation of pi Using Arithmetic-Geometric Mean ([[1939326/attachments/1341646|pdf]])<br>
 +
** E. Salamin, Mathematics of Computation 30(1976) 565-570

2009년 11월 17일 (화) 11:09 판

간단한 소개
  • AGM(arithmetic-geometric mean)을 활용하여 파이값을 빠르게 계산할 수 알고리즘

 

 

타원적분

\(K(k) = \int_0^{\frac{\pi}{2}} \frac{d\theta}{\sqrt{1-k^2 \sin^2\theta}}\)

 

\(E(k) = \int_0^{\frac{\pi}{2}} \sqrt{1-k^2 \sin^2\theta}}d\theta}{\)

\(q=e^{2\pi i \tau}\)

\(\theta_{2}(\tau)= \sum_{n=-\infty}^\infty q^{(n+\frac{1}{2})^2/2}\)

\(\theta_3(\tau)=\sum_{n=-\infty}^\infty q^{n^2/2}\)

\(\theta_{4}(\tau)= \sum_{n=-\infty}^\infty (-1)^n q^{n^2/2}\)

\(k=k(\tau)=\frac{\theta_2^2(\tau)}{\theta_3^2(\tau)}\)

\(K(k) = \int_0^{\frac{\pi}{2}} \frac{d\theta}{\sqrt{1-k^2 \sin^2\theta}}\)

\(E(k) = \int_0^{\frac{\pi}{2}} \sqrt{1-k^2 \sin^2\theta}}d\theta}{\)

\(k'=\sqrt{1-k^2}=\frac{\theta_4^2(\tau)}{\theta_3^2(\tau)}\)

\(K'(k) = K(k')\)

\(E'(k) = E(k')\)

 

 

타원적분에 대한 르장드르 항등식

For \(\phi\!\) and \(\theta\!\) such that \(\phi+\theta={1 \over 2}\pi\!\)

 

\[K(\sin \phi) E(\sin \theta ) + K(\sin \theta ) E(\sin \phi) - K(\sin \phi) K(\sin \theta) = {1 \over 2}\pi\!\]

  • 다음과 같이 표현 가능

\(E(k)K'(k)+E'(k)K(k)-K(k)K'(k)=\frac{\pi}{2}\)

특별히 다음과 같은 관계가 성립함

\(2K(\frac{1}{\sqrt{2}})E(\frac{1}{\sqrt{2}})-K(\frac{1}{\sqrt{2}})^2=\frac{\pi}{2}\)

 

 

타원적분과 AGM의 관계

\(K(k)=\frac{\pi}{2M(1,\sqrt{1-k^2})}\)

특별히, \(K(\frac{1}{\sqrt2})=\frac{\pi}{2M(1,\frac{1}{\sqrt2})}\)

 

한편, \(a_{n+1}={a_n+b_n \over 2}\),  \(b_{n+1}=\sqrt{a_n b_n}\) , \(a_0=1\), \(b_0=\sqrt{1-k^2}\) ,  \(c_n=\sqrt{a_n^2-b_n^2}\) 로 정의된 수열에 대하여, 타원적분은 다음과 같은 관계를 만족시킴.

 

\(\sum_{i=0}^{\infty} 2^{i-1} c_i^2 = 1 - \frac{E(k)}{K(k)}\)

 

 

가우스-살라민 알고리즘

[/pages/1939326/attachments/1341696 Salamin.jpg]

(증명)

\(2K(\frac{1}{\sqrt{2}})E(\frac{1}{\sqrt{2}})-K(\frac{1}{\sqrt{2}})^2=\frac{\pi}{2}\)

\(K(\frac{1}{\sqrt2})=\frac{\pi}{2M(1,\frac{1}{\sqrt2})}\)

 

\(\sum_{i=0}^{\infty} 2^{i-1} c_i^2 = 1 - \frac{E(k)}{K(k)}\)

를 결합하여 증명가능.

 

 

 

 

또다른 알고리즘

\(x_0=\sqrt{2}\) ,\(\pi_0=2+\sqrt{2}}\), \(y_1=\sqrt[4]{2}\)
\(x_n=\frac{1}{2}(\sqrt{x_n}+\frac{1}{\sqrt{x_n}})}, n\geq0\) , \(y_n=\frac{y_{n+1}\sqrt{x_n}+\frac{1}{\sqrt{x_{n}}}}{y_n+1}}, n\geq1\) , \(\pi_n=\pi_{n-1}\frac{x_n+1}{y_n+1}}, n\geq1\)

 

  • 위에 정의된 수열 \(\pi_n\)은 파이로 수렴하게 된다. 다음은 다섯번째 항까지 계산한 결과.

\(\pi_1=3.1426067539416226007907198236183018919713562462772\)
\(\pi_2=3.1415926609660442304977522351203396906792842568645\)
\(\pi_3=3.1415926535897932386457739917571417940347896238675\)
\(\pi_4=3.1415926535897932384626433832795028841972241204666\)
\(\pi_5=3.1415926535897932384626433832795028841971693993751\)

  • 한번씩 계산할 때마다, 대략 두 배 정도 정확한 자리수
  • 9번째까지 계산한다면, 1000자리 이상의 파이값을 계산
  • 매쓰매티카 노트

 

 

관련된 학부 과목과 미리 알고 있으면 좋은 것들

 

관련된 대학원 과목

 

 

관련된 다른 주제들

 

 

표준적인 도서 및 추천도서

 

 

위키링크

 

 

관련논문