1: \begin{abstract}

2: In this note, we consider  the Optimistic Gradient Ascent-Proximal Point Algorithm 

3: (OGAProx)

4:  proposed   by Bo{\c{t}}, Csetnek, and Sedlmayer for solving a saddle-point problem 

5:  associated with a convex-concave function 

6: constructed by a nonsmooth coupling function and one regularizing function.

7:  

8:  We first provide a counterexample to show that the convergence of the minimax gap

9:  function, evaluated at the ergodic sequences, is insufficient to demonstrate 

10:  the convergence of the function values evaluated at the ergodic 

11:  sequences. 

12:  Then under the same assumptions used by Bo{\c{t}} et al.\,for proving the

13:  convergence of the minimax gap function, 

14:   we present convergence results 

15:   for the function values evaluated at the ergodic sequences generated by the OGAProx

16:   with convergence rates of order 

17:   $\mathcal{O}\left(\frac{1}{k}\right)$, $\mathcal{O}\left(\frac{1}{k^{2}}\right)$,

18:   and $\mathcal{O}\left(\theta^{k}\right)$ with $\theta \in (0,1)$ 

19:   for the associated convex-concave coupling function being convex-concave, 

20:   convex-strongly concave, and strongly convex-strongly concave, respectively. 

21: \end{abstract}

22: