Inverse Problem techniques offer powerful tools which deal naturally with marginal data and asymmetric or strongly smoothing kernels, in cases where parameter-fitting methods may be used only with some caution. Although they are typically subject to some bias, they can invert data without requiring one to assume a particular model for the source. The Backus-Gilbert method in particular concentrates on the tradeoff between resolution and stability, and allows one to select an optimal compromise between them. We use these tools to analyse the problem of reconstructing features of the source star in a microlensing event, show that it should be possible to obtain useful information about the star with reasonably obtainable data, and note that the quality of the reconstruction is more sensitive to the number of data points than to the quality of individual ones.

It has already been shown that microlensing can give rise to a non-zero variable polarisation signal. Here we use realistic simulations to demonstrate the additional information that can be gained from polarimetric observations of lensing events.

We investigate the feasibility of reconstructing the radial intensity profile of extended stellar sources by inverting their microlensed light curves. Using a simple, linear, limb darkening law as an illustration, we show that the intensity profile can be accurately determined, at least over the outer part of the stellar disc, with realistic light curve sampling and photometric errors. The principal requirement is that the impact parameter of the lens be less than or equal to the stellar radius. Thus, the analysis of microlensing events provides a powerful method for testing stellar atmosphere models.

We describe a method for determining the limb polarization and limb darkening of stars in eclipsing binary systems, by inverting photometric and polarimetric light curves. Because of the ill-conditioning of the problem, we use the Backus-Gilbert method to control the resolution and stability of the recovered solution, and to make quantitative estimates of the maximum accuracy possible. Using this method we confirm that the limb polarization can indeed be recovered, and demonstrate this with simulated data, thus determining the level of observational accuracy required to achieve a given accuracy of reconstruction. This allows us to set out an optimal observational strategy, and to critcally assess the claimed detection of limb polarization in the Algol system. The use of polarization in stars has been proposed as a diagnostic tool in microlensing surveys by Simmons et al. (1995), and we discuss the extension of this work to the case of microlensing of extended sources.