diff --git a/Claudio_Maggioni_5/Claudio_Maggioni_5.md b/Claudio_Maggioni_5/Claudio_Maggioni_5.md index 31262e4..4866abf 100644 --- a/Claudio_Maggioni_5/Claudio_Maggioni_5.md +++ b/Claudio_Maggioni_5/Claudio_Maggioni_5.md @@ -121,12 +121,6 @@ central path C converges to anything then $\tau$ approaches zero, therefore leading the path to the constrained minimizer while keeping $x$ and $s$ positive but at the same time minimizing their pairwise products to zero. -In practice, most central path implementation use Newton steps toward points on -$\mathcal{C}$ for which $\tau > 0$, rather than pure Newton steps for $F$. This -is usually possible since those newton steps are biased to stay in the -hyperspace region where $x > 0$ and $s > 0$ even without enforcing these -constraints. - # Exercise 2 ## Exercise 2.1 @@ -213,27 +207,8 @@ constrained minimizer. # Exercise 3 ## Exercise 3.1 - - -Yes, the problem can be solved with _Uzzawa_'s method since the problem can be +Yes, the problem can be solved with Uzawa's method since the problem can be reformulated as a saddle point system. The KKT conditions of the problem can be reformulated as a matrix-vector to vector equality in the following way: @@ -256,7 +231,7 @@ g\\ h \end{bmatrix}$$ -This is the system the _Uzzawa_ method will solve. Therefore, we need to check +This is the system the Uzawa method will solve. Therefore, we need to check if the matrix: $$K = \begin{bmatrix}G & A^T \\ A& 0\end{bmatrix} = \begin{bmatrix} @@ -279,7 +254,7 @@ $$\begin{bmatrix} 8.7663\end{bmatrix}$$ Therefore, the system is indeed a saddle point system and it can be solved with -_Uzzawa_'s method. +Uzawa's method. ## Exercise 3.2 diff --git a/Claudio_Maggioni_5/Claudio_Maggioni_5.pdf b/Claudio_Maggioni_5/Claudio_Maggioni_5.pdf index e795b23..90eef9d 100644 Binary files a/Claudio_Maggioni_5/Claudio_Maggioni_5.pdf and b/Claudio_Maggioni_5/Claudio_Maggioni_5.pdf differ