1: \begin{abstract}

2: Douglas-Rachford Splitting (DRS) methods based on the proximal point algorithms for the Poisson and Gaussian log-likelihood functions 

3: are proposed for ptychography and phase retrieval.

4: 

5: 

6: 

7: Fixed point analysis shows  that the DRS iterated sequences are always bounded explicitly in terms of the step size

8: and that the fixed points are   attracting if and only if the fixed points are regular solutions.

9: This alleviates  two major drawbacks of the classical Douglas-Rachford algorithm:

10: slow convergence when the feasibility problem is consistent and divergent behavior when the feasibility problem is inconsistent. 

11: 

12: Fixed point analysis also leads to a simple, explicit expression for the optimal step size in terms of the spectral gap

13: of an underlying matrix. 

14: 

15: When applied to the challenging problem of {\em blind} ptychography, which seeks to recover both the object and the probe simultaneously,

16: Alternating Minimization with the DRS inner loops, even with  a far from optimal step size,  converges geometrically under the nearly minimum conditions established in the uniqueness theory. 

17: 

18:  

19: 

20: \commentout{Three boundary conditions are considered in the simulations: periodic, dark-field and bright-field boundary conditions.

21: The dark-field boundary condition is suited for isolated objects while the bright-field boundary condition is for non-isolated objects.

22: The periodic boundary condition is a mathematically convenient reference point. Depending on the availability of the boundary prior

23: the dark-field and 

24: the bright-field boundary conditions may or may not be enforced in the reconstruction.  Not surprisingly, enforcing the boundary condition 

25: improves the rate of convergence, sometimes in a significant way. Enforcing the bright-field condition in the reconstruction

26: can also remove

27: the linear phase ambiguity. 

28: }

29: 

30: \end{abstract}

31: