In Problem 1.7,

\begin{align*}
\prox_{h}(x_1,x_2)
&=
\frac{1}{2}\begin{bmatrix}
x_1+x_2+\prox_{2f}(x_1-x_2)\\
x_1+x_2+\prox_{2f}(x_1-x_2)
\end{bmatrix}.
\end{align*}

should be changed to

\begin{align*}
\prox_{h}(x_1,x_2)
&=
\frac{1}{2}\begin{bmatrix}
x_1+x_2+\prox_{2f}(x_1-x_2)\\
x_1+x_2-\prox_{2f}(x_1-x_2)
\end{bmatrix}.
\end{align*}


In Problem 2.28 equation (2.20), the minimization over x\in \reals^n should be changed to minimization over x,y\in \reals^n


The problem titled (PD3O generalizes DYS) was mistakenly not numbered. To avoid numbering inconsistencies between the online version and the printed book, I will number the exercise as 3.6-2.


In Problem 3.14 (Linearized method of multipliers and PDHG),
the g=\delta_{\{0\}} should be changed to g=\delta_{\{b\}}.

In Problem 3.15 (Chen--Teboulle is variable metric PPM) the diagonals $\alphaI$ of the metric matrix M should be changed to $\alpha^{-1}I$



In problem 3.17 (Unification of PAPC/PDFP$^2$O and Condat--V\~u), the 1,1 element of $M^{-1}$ is written as
\alpha I + \alpha^2\beta B^\intercal (I - \alpha\beta A^\intercal A)^{-1} B
this should be fixed to
\alpha I + \alpha^2\beta B^\intercal (I - \alpha\beta A A^\intercal)^{-1} B