Dirac Notation and Spectral decomposition Michele Mosca review”: Dirac notation t ψ For any vector ψ , we let denote ψ , the complex conjugate of ψ . We denote by φ ψ φ ψ the inner product between two vectors φ and ψ ψ defines a linear function that maps φ ψφ (I.e. ψ φ ψ φ … it maps any state φ to the coefficient of its ψ component) More Dirac notation ψ ψ defines a linear operator that maps ψ ψ φ ψ ψφ ψφ ψ This is a scalar so I can move it to front ψ (I.e. ψ projects a state to its ψ component Recall: this projection operator also corresponds to the “density matrix” for ψ ) More Dirac notation More generally, we can also have operators like θ ψ θ ψ φ θ ψφ ψφ θ Example of this Dirac notation For example, the one qubit NOT gate corresponds to the operator 0 1 1 0 e.g. 0 1 1 0 0 (sum of matrices applied to ket vector) 0 1 0 1 0 0 0 10 1 00 0 0 11 1 This is one more notation to calculate state from state and operator The NOT gate is a 1-qubit unitary operation. Special unitaries: Pauli Matrices in new notation The NOT operation, is often called the X or σX operation. 0 1 X X NOT 0 1 1 0 1 0 1 0 Z Z signflip 0 0 1 1 0 1 0 i Y Y i 0 1 i 1 0 i 0 Recall: Special unitaries: Pauli Matrices in new representation Representation of unitary operator What is e iHt ?? It helps to start with the spectral decomposition theorem. Spectral decomposition Definition: an operator (or matrix) M is t t “normal” if MM =M M t t E.g. Unitary matrices U satisfy UU =U U=I E.g. Density matrices (since they satisfy =t; i.e. “Hermitian”) are also normal Remember: Unitary matrix operators and density matrices are normal so can be decomposed Spectral decomposition Theorem Theorem: For any normal matrix M, there is a unitary matrix P so that M=PPt where is a diagonal matrix. The diagonal entries of are the eigenvalues. The columns of P encode the eigenvectors. Example: Spectral decomposition of the NOT gate X 0 1 X1 0 X 0 11 0 1 1 1 1 1 0 2 0 1 2 2 2 X { 0 , 1 } 1 1 1 1 1 0 0 1 2 2 2 2 1 1 1 1 0 1 0 1 2 2 2 2 X X X 1 0 X { , } 0 1 This is the middle matrix in above decomposition Spectral decomposition: matrix from column vectors a11 a 21 P an1 ψ1 a1n a2 n ann ψ2 ψn a12 a22 an 2 Column vectors Spectral decomposition: eigenvalues on diagonal λ1 Λ λ2 Eigenvalues on the diagonal λn Spectral decomposition: matrix as row vectors a a t P * a1n * 11 * 12 * 21 * 22 a a a2 n * a ψ1 ψ2 a ψ * ann n Adjont matrix = row vectors * n1 * n2 Spectral decomposition: using row and column vectors PΛP t ψ1 From theorem ψ2 λi ψi ψi i λ1 ψn λ1 0 0 λ2 0 λ2 0 ψ1 ψ 2 ψ λn n 0 th 0 i row i th colum n 0 0 0 λi 0 i λn 0 0 1 0 Verifying eigenvectors and eigenvalues PΛP t ψ 2 ψ1 ψ1 Multiply on right by state vector Psi-2 ψ2 ψ2 λ1 ψn λ1 ψn λ2 λ2 ψ1 ψ2 ψ λn n ψ2 ψ1 ψ2 ψ2 ψ2 ψ ψ λn n 2 Verifying eigenvectors and eigenvalues ψ1 ψ2 ψ1 ψ2 λ1 0 1 λ 2 ψ n 0 λn 0 λ2 ψ n λ2 ψ 2 0 useful Why is spectral decomposition useful? Because we can calculate f(A) m-th power Note that So ψ i ψi m ψi ψi recall ψi ψ j δij m m λi ψi ψi λi ψi ψi i i Consider f ( x) am x m m 1 e.g. e x m m m! x Why is spectral decomposition useful? Continue last slide M f M am M am λi ψi ψi m m i m m m am λi ψi ψi am λi ψi ψi m i i m f λi ψi ψi m i = f( i) Now f(M) will be in matrix notation f ( M ) am M m m f ( PΛP ) am PΛP t m λ1 P am m m t m m t am PΛ P P am Λ P m m m t m am λ1 m t P P m λn Pt m am λn m Same thing in matrix notation am λ1m m f ( PΛP t ) P f λ1 Pt P f λn ψ1 ψ2 Pt m m am λn f λ1 ψ n ψ1 ψ2 f λn ψn Same thing in matrix notation f λ1 f ( PΛP t ) P ψ1 ψ2 f λi ψi ψi i Pt f λn f λ1 ψ n ψ1 ψ2 f λn ψn Important formula in matrix notation f ( PP ) f i i i t i “Von Neumann measurement in the computational basis” Suppose we have a universal set of quantum gates, and the ability to measure each qubit in the basis { 0 , 1 } If we measure (0 0 1 1 ) we get 2 with probability b α b We knew it from beginning but now we can generalize Using new notation this can be described like this: We have the projection operators P0 0 0 and P1 1 1 satisfying P0 P1 I We consider the projection operator or “observable” M 0P0 1P1 P1 Note that 0 and 1 are the eigenvalues When we measure this observable M, the probability of getting the eigenvalue b is 2 Pr( b) Φ Pb Φ α b and we are in that case left with the state Pb b b b p(b) b Polar Decomposition Left polar decomposition Right polar decomposition This is for square matrices Gram-Schmidt Orthogonalization Hilbert Space: Orthogonality: Norm: Orthonormal basis: