1: \begin{abstract}

2: Future experiments will produce high-resolution temperature maps of the

3: cosmic microwave background (CMB) and are expected to reveal the signature

4: of gravitational lensing by intervening large-scale structures.  We

5: construct all-sky maximum-likelihood estimators that use the lensing

6: effect to estimate the projected density (convergence) of these

7: structures, its power spectrum, and cross-correlation with other

8: observables.  This contrasts with earlier quadratic-estimator approaches

9: that Taylor-expanded the observed CMB temperature to linear order in the

10: lensing deflection angle; these approaches gave estimators for the

11: temperature-convergence correlation in terms of the CMB three-point

12: correlation function and for the convergence power spectrum in terms of

13: the CMB four-point correlation

14: function, which can be biased and non-optimal due to terms beyond the

15: linear order.  

16: We show that for sufficiently weak lensing, the maximum-likelihood

17: estimator reduces to the computationally less

18: demanding quadratic estimator.  The maximum likelihood and quadratic

19: approaches are compared by evaluating the

20: root-mean-square (RMS) error and bias in the reconstructed convergence

21: map in a numerical simulation; 

22: it is found that both the RMS errors and bias 

23: are of order 1 percent for the case of Planck and of order 10--20 percent

24: for a 1 arcminute beam 

25: experiment.  We

26: conclude that for recovering lensing information from temperature data

27: acquired by these experiments, the quadratic

28: estimator is close to optimal, but

29: further work will be required to determine whether this is also the case

30: for lensing of the CMB polarization field.

31: \end{abstract}

32: