How to Prove that a Matrix is Positive Definite

preview_player
Показать описание
In this video I will teach you what a positive definite matrix is and how you can prove that a matrix is positive definite using the five fundamental properties of positive definite matrices. I will also show you a worked example of using Sylvester's criteria for positive definite matrices to actually prove that a matrix is in fact positive definite. I will also share with you a really simple way to instantly tell by looking at a matrix if it is likely to be positive definite or if it cannot be positive definite.
Рекомендации по теме
Комментарии
Автор

Great explanation! Thank you! However, to the best of my knowledge, the Sylvester's criterion is a necessary and sufficient condition only for the symmetric matrices (or Hermitian matrices, if we include complex numbers). The final example was symmetric, but not the ones before that.

markmisin
Автор

I thought the +ve eigenvalues rule only works if your matrix is symmetric

kaiochanx
Автор

thank you very much, it was very useful, Allah bless you.

zinebadaika
Автор

@1:23 That is a weird definition of "symmetric". Usually, it's defined as A_(i, j) = A_(j, i), which seems like a way more natural way to define it. From that, I guess one can prove that there is some other matrix B such that A = B^T*B, but I don't think I've ever seen that before.

Furthermore, I think 4 can be stated much more simply as "the matrix is symmetric (and full rank)"

allyourcode
Автор

this is so useful, thanks for sharing the video!

whogashaga
Автор

I wish he would of explained a couple things.

1) What if there was a different number other than zero?
2) How was the last matrix created that have -1, -1, 0, & 2?

investwithvincent