This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Problem Set #7 Solutions Clifford Algebras For the first 4 problems, let R be a ring with 1 and let M n ( R ) be the ring of n n matrices with entries in R . The set R n of column vectors of length n is then an ( M n ( R ) , R )bimodule, and its transpose t R n is an ( R, M n ( R )) bimodule. 1. Show that any A M n ( R ) can be written as a sum of rank 1 matrices. Let A M n ( R ) and let E ij M n ( R ) denote the matrix with a 1 in the ( i, j )th slot and zeros elsewhere. Clearly the Rdimension of the image of each E ij acting on R n on the left or on t R n on the right is 1, so in any case E ij is a rank 1 matrix. Letting a ij R be the ( i, j )th entry of A we have A = X 1 i,j n a ij E ij . 2. Given A M n ( R ) , show that there exists v 1 , . . . , v k R n and w t 1 , . . . , w t k t R n such that M = 1 i k v i w t i . Let A M n ( R ) with entries a ij R and let e i R n have a 1 in the i th slot and zeros elsewhere. Then since e i e t j = E ij it follows from part 1 that A = X 1 i,j n a ij e i e t j . Taking v j = i a ij e i and w j = e j we obtain our desired result. 3. Conclude that R n R t R n = M n ( R ) , an isomorphism of ( M n ( R ) , M n ( R )) bimodules. Note that the e i , 1 i n , form an Rbasis for R n , so R n R t R n has basis e i e t j , 1 i, j n . Define : R n R t R n M n ( R ) by set ting ( e i e t j ) = e i e t j = E ij on this basis and extending ( R, R )bilinearly to R n R t R n . Let A M n ( R ) with entries a s` R ; then by part b, ( Ae i e t j ) = 1 s,` n a s` e s e t ` e i e t j = 1 s,` n a s` `,i e s e t j = 1 s n a si e s e t j = 1 s n a si ( e s e t j ) = 1 s n a si E sj = 1 s,` n a s` `,i E sj = 1 s,` n a s` E s` E ij = 1 s,` n a s` E s` ( e i e t j ) = A ( e i e t j ) , 1 and by symmetry ( e i e t j A ) = ( e i e t j ) A . We check is welldefined: Let r, a i , b j R . Then using exercise IV.6.5 and the ( R, R )linearity of we find i ( a i e i ) r j e t j b j = i,j a i ( ( re i ) e t j ) b j = i,j a i ( r ( e i e t j )) b j = i,j a i r ( e i e t j ) b j = i,j a i rE ij b j = i,j a i E ij rb j = i,j a i ( e i e t j ) rb j = i,j a i (( e i e t j ) r ) b j = i,j a i ( e i ( e t j r )) b j = i a i e i r j e t j b j ....
View
Full
Document
 Spring '08
 MORRISON
 Algebra, Vectors, Matrices

Click to edit the document details