Householder Transformation with QR Decomposition Examples.

แชร์
ฝัง
  • เผยแพร่เมื่อ 30 ก.ย. 2024
  • Householder Reflection or Transformation is one the methods of decomposing a matrix into an Orthogonal Matrix (Q) and Right Upper Triangular Matrix (R). It helps to solve systems of equation using backward substitution.

ความคิดเห็น • 44

  • @veenanarayanan9490
    @veenanarayanan9490 ปีที่แล้ว +5

    Thank you for the explanation sir. In the first example, the signs of all the elements in the second column of Q2 is positive. The first entry of the third column of Q2 is positive and the sign of third entry is negative. i.e Q2= H1*H2=[ -0.5774 0.4083 0.7071; 0.5774 0.8165 -0.0000; -0.5774 0.4082 -0.7071]. Therefore, Q2*R2=B.

  • @MoumitaMandal-d3p
    @MoumitaMandal-d3p 4 วันที่ผ่านมา

    Q2= (H2 H1)^ T , Transpose is missing. Though here both are same

  • @ae00505
    @ae00505 25 วันที่ผ่านมา

    Thanks man. in the book it was not clear to me. you helped me a lot. thanks.

  • @jcf129er
    @jcf129er 7 หลายเดือนก่อน +3

    Just wanted to say your whole channel, numerical methods + examples, is absolutely solid! Thank you so much for the hard work!

  • @MrNike799
    @MrNike799 3 ปีที่แล้ว +7

    you should mention that u1 = a1 + sign(a_11)||a1||e1. It's not always minus.

  • @HKHasty
    @HKHasty 7 หลายเดือนก่อน +1

    Very helpful, how come though you aren't taking the sign of first element in x to calculate v?
    v_k = x+sign(x_k) ||x||_2 e_k, with k being column number on original matrix B.

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  7 หลายเดือนก่อน

      In practice, both choices v=a+∥a∥e1 and v=a−∥a∥e1 work. You can try solving Question one with both signs and see the difference. I believe it will work. But if v will be zero then you need to checking the signs so mostly this is consider v=a+sign(a_11)∥a∥e1. Thanks

  • @rjk1.618
    @rjk1.618 10 หลายเดือนก่อน +2

    Super helpful, first clear and concise video explaining it I found. Thank you !!

  • @planetariansaturn8593
    @planetariansaturn8593 9 หลายเดือนก่อน +1

    This is amazing help with linear algebra, I thank you so much helped a lot

  • @kingshukdutta2064
    @kingshukdutta2064 ปีที่แล้ว +2

    fucktorization lmao. love your accent

  • @AidanHomewood
    @AidanHomewood ปีที่แล้ว +2

    Bro you are incredible. Best explanation I've heard after ages of trying to understand how to do this. Keep it up man.

  • @mb79c.57
    @mb79c.57 ปีที่แล้ว +2

    Great explanation, thank you very much!!

  • @ioannisvarnas9641
    @ioannisvarnas9641 2 ปีที่แล้ว +2

    Such an amazing explanation
    Good Job !!!

  • @georgeklimov3464
    @georgeklimov3464 ปีที่แล้ว +1

    Thank you for the video, I learned a lot. I would like to point out that there is a mistake on the second problem, I will not explain it here but Q = P1P2 has a calculation error. Otherwise, phenomenal video!

  • @giorgioconiglio5116
    @giorgioconiglio5116 6 หลายเดือนก่อน +1

    Thank you, great job!

  • @founderinho
    @founderinho 2 ปีที่แล้ว +1

    What should we do with complex matrices?

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  2 ปีที่แล้ว

      Actually, I have not worked on any problem having complex matrix. But I believe Householder Transformation should be able to handle it. Thanks.

  • @twominQuotes
    @twominQuotes 7 หลายเดือนก่อน +1

    Thank you so much

  • @ryanjackson0x
    @ryanjackson0x 2 ปีที่แล้ว +1

    Doesn't x1 itself usually come from a matrix? Can you show or explain that?

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  2 ปีที่แล้ว

      x1 is from the matrix and and the subsequent x's are from the R's matrices. I believe the explanation is helpful.

    • @nguyetmainguyenthi2292
      @nguyetmainguyenthi2292 2 ปีที่แล้ว

      @@isaacamornorteyyowetu how about x2?

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  2 ปีที่แล้ว +1

      @@nguyetmainguyenthi2292 The subsequent x's are from the R's matrices.

  • @kaushikbhanja7747
    @kaushikbhanja7747 ปีที่แล้ว +1

    thanks for the discussion

  • @singletaxjax307
    @singletaxjax307 2 ปีที่แล้ว +1

    Thank you for the help, sir

  • @ricardoraymond9037
    @ricardoraymond9037 2 ปีที่แล้ว

    Nathan kutz covered the same topic, and only he and god knows what he is talking about.

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  2 ปีที่แล้ว

      It is nice that this video has been helpful to you. Thanks for the feedback.

  • @bhageshmaheshwari2807
    @bhageshmaheshwari2807 3 ปีที่แล้ว +2

    In the two examples sign of first element was different but you still used - ? Is that a mistake ?

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  3 ปีที่แล้ว +1

      The first one was - and the second one was +. It wasn't a mistake. In practice, both choices v=a+∥a∥e1 and v=a−∥a∥e1 work. You can try solving Question one with both signs and see the difference. I believe it will work. But if v will be zero then you need to checking the signs so mostly this is consider v=a+sign(a_11)∥a∥e1. Thanks

    • @isaacamornorteyyowetu
      @isaacamornorteyyowetu  2 ปีที่แล้ว

      @Haonan Yuan I think a lot needs to be studied with reference to the sign to use. But in all the examples, i maintained just one sign(-) and it worked. Kindly do a further check and help me to know on your discovery. Thanks

    • @kingshukdutta2064
      @kingshukdutta2064 ปีที่แล้ว +1

      @@isaacamornorteyyowetu so in 2nd case it will be positive right?