linear algebra and differential equations math 54 lecture
play

Linear algebra and differential equations (Math 54): Lecture 13 - PowerPoint PPT Presentation

Linear algebra and differential equations (Math 54): Lecture 13 Vivek Shende March 7, 2019 Hello and welcome to class! Hello and welcome to class! Last time Hello and welcome to class! Last time We began discussing eigenvalues and


  1. Example  0 1 0  Consider the matrix 0 0 1  .  1 0 0  − λ 1 0   = 1 − λ 3 Its characteristic equation is: 0 = det 0 − λ 1  1 0 − λ This has one real solution: λ = 1.   − 1 1 0 Eigenvectors of eigenvalue 1 form the kernel of 0 − 1 1  .  1 0 − 1 This is spanned by (1 , 1 , 1).

  2. Example  0 1 0  Consider the matrix 0 0 1  .  1 0 0  − λ 1 0   = 1 − λ 3 Its characteristic equation is: 0 = det 0 − λ 1  1 0 − λ This has one real solution: λ = 1.   − 1 1 0 Eigenvectors of eigenvalue 1 form the kernel of 0 − 1 1  .  1 0 − 1 This is spanned by (1 , 1 , 1). There is no basis of real eigenvectors.

  3. Example � � 0 1 Consider the matrix . − 1 0

  4. Example � � 0 1 Consider the matrix . − 1 0 Its characteristic equation is:

  5. Example � � 0 1 Consider the matrix . − 1 0 � − λ � 1 = λ 2 + 1. This Its characteristic equation is: 0 = det − 1 − λ has

  6. Example � � 0 1 Consider the matrix . − 1 0 � − λ � 1 = λ 2 + 1. This Its characteristic equation is: 0 = det − 1 − λ has no real solutions, so there are no real eigenvalues or nonzero eigenvectors.

  7. Example � � 0 1 Consider the matrix . − 1 0 � − λ � 1 = λ 2 + 1. This Its characteristic equation is: 0 = det − 1 − λ has no real solutions, so there are no real eigenvalues or nonzero eigenvectors. There is no basis of real eigenvectors.

  8. Example � 1 � 1 Consider the matrix . 0 1

  9. Example � 1 � 1 Consider the matrix . 0 1 Characteristic equation:

  10. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ

  11. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are

  12. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1

  13. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 The eigenvectors of eigenvalue 1

  14. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 1 The eigenvectors of eigenvalue 1 are the kernel of . 0 0

  15. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 1 The eigenvectors of eigenvalue 1 are the kernel of . This 0 0 is spanned by

  16. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 1 The eigenvectors of eigenvalue 1 are the kernel of . This 0 0 is spanned by (1 , 0).

  17. Example � 1 � 1 Consider the matrix . 0 1 � 1 − λ � 1 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 1 The eigenvectors of eigenvalue 1 are the kernel of . This 0 0 is spanned by (1 , 0). There is no basis of real eigenvectors.

  18. Example � 1 � 0 Consider the matrix . 0 1

  19. Example � 1 � 0 Consider the matrix . 0 1 Characteristic equation:

  20. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ

  21. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are

  22. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1

  23. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 The eigenvectors of eigenvalue 1

  24. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 0 The eigenvectors of eigenvalue 1 are the kernel of . 0 0

  25. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 0 The eigenvectors of eigenvalue 1 are the kernel of . This 0 0 is all vectors in R 2 .

  26. Example � 1 � 0 Consider the matrix . 0 1 � 1 − λ � 0 = λ 2 − 2 λ + 1. Characteristic equation: 0 = det 0 1 − λ So the eigenvalues are λ = 1 � 0 � 0 The eigenvectors of eigenvalue 1 are the kernel of . This 0 0 is all vectors in R 2 . There is a basis of real eigenvectors.

  27. Example d Consider the linear transformation dx : P n → P n .

  28. Example d Consider the linear transformation dx : P n → P n . Constant polynomials are eigenvectors of eigenvalue zero.

  29. Example d Consider the linear transformation dx : P n → P n . Constant polynomials are eigenvectors of eigenvalue zero. There are no others:

  30. Example d Consider the linear transformation dx : P n → P n . Constant polynomials are eigenvectors of eigenvalue zero. There are no others: the derivative lowers the degree of a polynomial

  31. Example d Consider the linear transformation dx : P n → P n . Constant polynomials are eigenvectors of eigenvalue zero. There are no others: the derivative lowers the degree of a polynomial so any eigenvector must have eigenvalue zero

  32. Example d Consider the linear transformation dx : P n → P n . Constant polynomials are eigenvectors of eigenvalue zero. There are no others: the derivative lowers the degree of a polynomial so any eigenvector must have eigenvalue zero hence be a constant.

  33. Example Or, we pick a basis, say 1 , x , x 2 , . . .

  34. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx .

  35. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :

  36. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0

  37. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0 The characteristic equation of this matrix is − λ 5 = 0.

  38. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0 The characteristic equation of this matrix is − λ 5 = 0. So 0 is the only eigenvalue.

  39. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0 The characteristic equation of this matrix is − λ 5 = 0. So 0 is the only eigenvalue. The corresponding eigenspace is

  40. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0 The characteristic equation of this matrix is − λ 5 = 0. So 0 is the only eigenvalue. The corresponding eigenspace is in this basis

  41. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0 The characteristic equation of this matrix is − λ 5 = 0. So 0 is the only eigenvalue. The corresponding eigenspace is in this basis spanned by (1 , 0 , 0 , 0 , 0).

  42. Example Or, we pick a basis, say 1 , x , x 2 , . . . and write the matrix of d dx . I’ll write the case P 4 :  0 1 0 0 0  0 0 2 0 0     0 0 0 3 0     0 0 0 0 4   0 0 0 0 0 The characteristic equation of this matrix is − λ 5 = 0. So 0 is the only eigenvalue. The corresponding eigenspace is in this basis spanned by (1 , 0 , 0 , 0 , 0). These are the constant polynomials.

  43. When is there a basis of eigenvectors?

  44. When is there a basis of eigenvectors? Theorem Let T : V → V be a linear transformation,

  45. When is there a basis of eigenvectors? Theorem Let T : V → V be a linear transformation, and suppose v 1 , . . . , v n are nonzero eigenvectors, T v i = λ i v i

  46. When is there a basis of eigenvectors? Theorem Let T : V → V be a linear transformation, and suppose v 1 , . . . , v n are nonzero eigenvectors, T v i = λ i v i If the eigenvalues λ i are distinct, then the eigenvectors v i are linearly independent.

  47. When is there a basis of eigenvectors? Proof.

  48. When is there a basis of eigenvectors? Proof. We check n = 2.

  49. When is there a basis of eigenvectors? Proof. We check n = 2. Assume v 1 , v 2 are linearly dependent, i.e. c 1 v 1 + c 2 v 2 = 0 where c 1 , c 2 are constants not both zero. Applying T , we have λ 1 c 1 v 1 + λ 2 c 2 v 2 = 0 as well.

  50. When is there a basis of eigenvectors? Proof. We check n = 2. Assume v 1 , v 2 are linearly dependent, i.e. c 1 v 1 + c 2 v 2 = 0 where c 1 , c 2 are constants not both zero. Applying T , we have λ 1 c 1 v 1 + λ 2 c 2 v 2 = 0 as well. If c 1 = 0,

  51. When is there a basis of eigenvectors? Proof. We check n = 2. Assume v 1 , v 2 are linearly dependent, i.e. c 1 v 1 + c 2 v 2 = 0 where c 1 , c 2 are constants not both zero. Applying T , we have λ 1 c 1 v 1 + λ 2 c 2 v 2 = 0 as well. If c 1 = 0, then we can conclude v 2 = 0,

  52. When is there a basis of eigenvectors? Proof. We check n = 2. Assume v 1 , v 2 are linearly dependent, i.e. c 1 v 1 + c 2 v 2 = 0 where c 1 , c 2 are constants not both zero. Applying T , we have λ 1 c 1 v 1 + λ 2 c 2 v 2 = 0 as well. If c 1 = 0, then we can conclude v 2 = 0, which is a contradiction since we assumed the v i � = 0.

  53. When is there a basis of eigenvectors? Proof. We check n = 2. Assume v 1 , v 2 are linearly dependent, i.e. c 1 v 1 + c 2 v 2 = 0 where c 1 , c 2 are constants not both zero. Applying T , we have λ 1 c 1 v 1 + λ 2 c 2 v 2 = 0 as well. If c 1 = 0, then we can conclude v 2 = 0, which is a contradiction since we assumed the v i � = 0. So c 1 � = 0.

  54. When is there a basis of eigenvectors? Proof. We check n = 2. Assume v 1 , v 2 are linearly dependent, i.e. c 1 v 1 + c 2 v 2 = 0 where c 1 , c 2 are constants not both zero. Applying T , we have λ 1 c 1 v 1 + λ 2 c 2 v 2 = 0 as well. If c 1 = 0, then we can conclude v 2 = 0, which is a contradiction since we assumed the v i � = 0. So c 1 � = 0. If λ 2 = 0, then λ 1 c 1 v 1 = 0

Recommend


More recommend