1: \begin{abstract}

2: This paper investigates optimal error bounds and convergence rates for general Mann iterations for computing

3:  fixed-points of non-expansive maps in normed spaces. We look for iterations that achieve the smallest fixed-point 

4: residual after $n$ steps, by  minimizing a worst-case bound $\|x^n-Tx^n\|\le R_n$ derived from a nested family of 

5: optimal transport problems. We prove that this bound is tight so that minimizing $R_n$ yields optimal  

6: iterations. Inspired from numerical results we identify iterations that attain the rate $R_n=O(1/n)$, 

7: which we also show to be the best possible.

8: In particular, we prove that the classical Halpern iteration achieves this optimal rate

9: for several alternative stepsizes, and we determine analytically the optimal stepsizes that 

10: attain the smallest worst-case residuals at every step $n$,

11: with a tight bound $R_n\approx\frac{4}{n+4}$. 

12: We also determine the optimal Halpern stepsizes for

13: affine nonexpansive maps, for which we get exactly $R_n=\frac{1}{n+1}$. 

14: Finally, we show that the best rate for the classical Krasnosel'ski\u{\i}-Mann iteration is 

15: $\Omega(1/\sqrt{n})$, and we present numerical evidence suggesting that even after

16: introducing inertial terms one cannot reach the faster rate $O(1/n)$.

17: \end{abstract}

18: