A gravitational lens refers to a distribution of matter (such as a cluster of galaxies) between a distant light source and an observer, that's capable of bending the light from the source as the light travels towards the observer. This effect is known as gravitational lensing, and the amount of bending is one of the predictions of Albert Einstein's general theory of relativity.   (Classical physics additionally predicts the bending of light, but only half that predicted by general relativity.)
Although either Orest Khvolson (1924) or Frantisek Link (1936) is at times credited as being the first to discuss the effect in print, the effect is more commonly associated with Einstein, who published a more famous article on the subject in 1936.
Fritz Zwicky posited in 1937 that the effect could allow galaxy clusters to act as gravitational lenses. It wasn't until 1979 that this effect was confirmed by observation of the so-called " Twin QSO " SBS 0957+561.
Unlike an optical lens, a gravitational lens bends to the maximum light that passes closest to its center, and to a minimum light that travels furthest from its center. Consequently, a gravitational lens has no single focal point, but a focal line. The term "lens" in the context of gravitational light deflection was first used by O.J. Lodge, who remarked that it is "not permissible to say that the solar gravitational field acts like a lens, for it has no focal length". If the (light) source, the massive lensing object, and the observer lie in a straight line, the original light source will appear as a ring around the massive lensing object. If there's any misalignment, the observer will see an arc segment instead. This phenomenon was first mentioned in 1924 by the St. Petersburg physicist Orest Chwolson,  and quantified by Albert Einstein in 1936. It is most of the time referred to in the literature as an Einstein ring, after Chwolson didn't concern himself with the flux or radius of the ring image. More commonly, where the lensing mass is complex (such as a galaxy group or cluster) and doesn't cause a spherical distortion of space–time, the source will resemble partial arcs scattered around the lens. The observer might then see multiple distorted images of the same source; the number and shape of these depending upon the relative positions of the source, lens, and observer, and the shape of the gravitational well of the lensing object. 
There are three classes of gravitational lensing:
1. Strong lensing: where there are easily visible distortions like the formation of Einstein rings, arcs, and multiple images.
2. Weak lensing: where the distortions of background sources are much smaller and can only be detected by analysing large numbers of sources in a statistical way to find coherent distortions of only a few percent. The lensing shows up statistically as a preferred stretching of the background objects perpendicular to the direction to the centre of the lens. By measuring the shapes and orientations of large numbers of distant galaxies, their orientations can be averaged to measure the shear of the lensing field in any region. This, in turn, can be used to reconstruct the mass distribution in the area: in particular, the background distribution of dark matter can be reconstructed. Since galaxies are intrinsically elliptical and the weak gravitational lensing signal is small, a quite large number of galaxies must be used in these surveys. These weak lensing surveys must carefully avoid a number of important sources of systematic error: the intrinsic shape of galaxies, the tendency of a camera's point spread function to distort the shape of a galaxy and the tendency of atmospheric seeing to distort images must be understood and carefully accounted for. The results of these surveys are important for cosmological parameter estimation, to better understand and improve upon the Lambda-CDM model, and to provide a consistency cheque on additional cosmological observations. They might additionally provide an important future constraint on dark energy.
3. Microlensing: where no distortion in shape can be seen but the amount of light received from a background object changes in time. The lensing object might be stars in the Milky Way in one typical case, with the background source being stars in a remote galaxy, or, in another case, an even more distant quasar. The effect is small, such that (in the case of strong lensing) even a galaxy with a mass more than 100 billion times that of the Sun will produce multiple images separated by only a few arcseconds. Galaxy clusters can produce separations of several arcminutes. In both cases the galaxies and sources are quite distant, a large number of hundreds of megaparsecs away from our Galaxy.
Gravitational lenses act equally on all kinds of electromagnetic radiation, not just visible light. Weak lensing effects are being studied for the cosmic microwave background as well as galaxy surveys. Strong lenses have been observed in radio and x-ray regimes as well. If a strong lens produces multiple images, there will be a relative time delay between two paths: that is, in one image the lensed object will be observed before the additional image.
Henry Cavendish in 1784 (in an unpublished manuscript) and Johann Georg von Soldner in 1801 (published in 1804) had pointed out that Newtonian gravity predicts that starlight will bend around a massive object as had already been supposed by Isaac Newton in 1704 in his famous Queries No.1 in his book Opticks. The same value as Soldner's was calculated by Einstein in 1911 based on the equivalence principle alone. Notwithstanding Einstein noted in 1915, in the process of completing general relativity, that his (and thus Soldner's) 1911-result is only half of the correct value. Einstein became the first to calculate the correct value for light bending. 
The first observation of light deflection was performed by noting the change in position of stars as they passed near the Sun on the celestial sphere. The observations were performed in May 1919 by Arthur Eddington, Frank Watson Dyson, and their collaborators throughout a total solar eclipse.  The solar eclipse allowed the stars near the Sun to be observed. Observations were made simultaneously in the cities of Sobral, Ceará, Brazil and in São Tomé and Príncipe on the west coast of Africa. The observations demonstrated that the light from stars passing close to the Sun was slightly bent, so that stars appeared slightly out of position. 
The result was considered spectacular news and made the front page of most major newspapers. It made Einstein and his theory of general relativity world-famous. When asked by his assistant what his reaction would have been if general relativity hadn't been confirmed by Eddington and Dyson in 1919, Einstein famously made the quip: "Then I would feel sorry for the dear Lord. The theory is correct anyway."
Even before his breakthrough in the formulation of general relativity, Einstein realised that due to light deflection it was additionally possible that a mass could deflect light along two different paths causing the observer to see multiple images of a single source; this effect would make the mass act as a kind of gravitational lens. Notwithstanding as he only considered the effect in relation to single stars, he seemed to conclude that the phenomenon was unlikely to be observed for the foreseeable future after the necessary alignments between stars and observer would be highly improbable. Several additional physicists speculated about gravitational lensing as well, but all reached the same conclusion that it would be nearly impossible to observe. 
In 1936, after a few urging by Rudi W. Mandl, Einstein reluctantly published the short article "Lens-Like Action of a Star By the Deviation of Light In the Gravitational Field" in the journal Science . 
In 1937, Fritz Zwicky first considered the case where the newly detected galaxies (which were called 'nebulae' at the time) could act as both source and lens, and that, because of the mass and sizes involved, the effect was much more likely to be observed. 
It wasn't until 1979 that the first gravitational lens would be discovered. It became known as the " Twin QSO " after it initially looked like two identical quasistellar objects. (It is officially named SBS 0957+561.) This gravitational lens was detected by Dennis Walsh, Bob Carswell, and Ray Weymann using the Kitt Peak National Observatory 2.1 metre telescope. 
In the 1980s, astronomers realised that the combination of CCD imagers and computers would allow the brightness of millions of stars to be measured each night. In a dense field, like the galactic centre or the Magellanic clouds, a large number of microlensing events per year could potentially be found. This led to efforts like Optical Gravitational Lensing Experiment, or OGLE, that have characterised hundreds of such events.
Explanation in terms of space–time curvature
In general relativity, light follows the curvature of spacetime, hence when light passes around a massive object, it is bent. This means that the light from an object on the additional side will be bent towards an observer's eye, just like an ordinary lens. Since light always moves at a constant speed, lensing changes the direction of the velocity of the light, but not the magnitude.
Light rays are the boundary between the future, the spacelike, and the past regions. The gravitational attraction can be viewed as the motion of undisturbed objects in a background curved geometry or alternatively as the response of objects to a force in a flat geometry. The angle of deflection is:
toward the mass M at a distance r from the affected radiation, where G is the universal constant of gravitation and c is the speed of light in a vacuum.
Since the Schwarzschild radius is defined as , this can additionally be expressed in simple form as
Search for gravitational lenses
Most of the gravitational lenses in the past have been detected accidentally. A search for gravitational lenses in the northern hemisphere (Cosmic Lens All Sky Survey, CLASS), done in radio frequencies using the Very Large Array (VLA) in New Mexico, led to the discovery of 22 new lensing systems, a major milestone. This has opened a whole new avenue for research ranging from finding quite distant objects to finding values for cosmological parameters so we can understand the universe better.
A similar search in the southern hemisphere would be a quite good step towards complementing the northern hemisphere search as well as obtaining additional objectives for study. If such a search is done using well-calibrated and well-parameterized instrument and data, a result similar to the northern survey can be expected. The use of the Australia Telescope 20 GHz (AT20G) Survey data collected using the Australia Telescope Compact Array (ATCA) stands to be such a collection of data. As the data were collected using the same instrument maintaining a quite stringent quality of data we should expect to obtain good results from the search. The AT20G survey is a blind survey at 20 GHz frequency in the radio domain of the electromagnetic spectrum. Due to the high frequency used, the chances of finding gravitational lenses increases as the relative number of compact core objects (e.g. Quasars) are higher (Sadler et al. 2006). This is important as the lensing is easier to detect and identify in simple objects compared to objects with complexity in them. This search involves the use of interferometric methods to identify candidates and follow them up at higher resolution to identify them. Full detail of the project is currently under works for publication.
In a 2009 article on Science Daily a team of scientists led by a cosmologist from the U.S. Department of Energy's Lawrence Berkeley National Laboratory has made major progress in extending the use of gravitational lensing to the study of much older and smaller structures than was previously possible by stating that weak gravitational lensing improves measurements of distant galaxies. 
Astronomers from the Max Planck Institute for Astronomy in Heidelberg, Germany, the results of which are accepted for publication on Oct 21, 2013 in the Astrophysical Journal Letters (arXiv.org), detected what at the time was the most distant gravitational lens galaxy termed as J1000+0221 using NASA ’s Hubble Space Telescope.  While it remains the most distant quad-image lensing galaxy known, an even more distant two-image lensing galaxy was subsequently detected by an international team of astronomers using a combination of Hubble Space Telescope and Keck telescope imaging and spectroscopy. The discovery and analysis of the IRC 0218 lens was published in the Astrophysical Journal Letters on June 23, 2014.
Research published Sep 30, 2013 in the online edition of Physical Review Letters, led by McGill University in Montreal, Québec, Canada, has detected the B-modes, that are formed due to gravitational lensing effect, using National Science Foundation's South Pole Telescope and with help from the Herschel space observatory. This discovery would open the possibilities of testing the theories of how our universe originated. 
Solar gravitational lens
Albert Einstein predicted in 1936 that rays of light from the same direction that skirt the edges of the Sun would converge to a focal point approximately 542 AU from the Sun.  Thus, a probe positioned at this distance (or greater) from the Sun could use the sun as a gravitational lens for magnifying distant objects on the opposite side of the sun A probe's location could shift around as needed to select different targets relative to the Sun.
This distance is far beyond the progress and equipment capabilities of space probes like Voyager 1, and beyond the known planets and dwarf planets, though over thousands of years 90377 Sedna will move further away on its highly elliptical orbit. The high gain for potentially detecting signals through this lens, like microwaves at the 21-cm hydrogen line, led to the suggestion by Frank Drake in the early days of SETI that a probe can be sent to this distance. A multipurpose probe SETISAIL and later FOCAL was proposed to the ESA in 1993, but is expected to be a difficult task.  If a probe does pass 542 AU, magnification capabilities of the lens will continue to act at further distances, as the rays that come to a focus at larger distances pass further away from the distortions of the Sun's corona. 
Measuring weak lensing
Kaiser et al. (1995), Luppino & Kaiser (1997) and Hoekstra et al. (1998) prescribed a method to invert the effects of the Point Spread Function (PSF) smearing and shearing, recovering a shear estimator uncontaminated by the systematic distortion of the PSF. This method (KSB+) is the most widely used method in current weak lensing shear measurements.
Galaxies have random rotations and inclinations. As a result, the shear effects in weak lensing need to be determined by statistically preferred orientations. The primary source of error in lensing measurement is due to the convolution of the PSF with the lensed image. The KSB method measures the ellipticity of a galaxy image. The shear is proportional to the ellipticity. The objects in lensed images are parameterized according to their weighted quadrupole moments. For a perfect ellipse, the weighted quadrupole moments are related to the weighted ellipticity. KSB calculate how a weighted ellipticity measure is related to the shear and use the same formalism to remove the effects of the PSF.
KSB’s primary advantages are its mathematical ease and relatively simple implementation. Notwithstanding KSB is based on a key assumption that the PSF is circular with an anisotropic distortion. It’s fine for current cosmic shear surveys, but the next generation of surveys (e.g. LSST) might need much better accuracy than KSB can provide. Because throughout that time, the statistical errors from the data are negligible, the systematic errors will dominate.