"Theoretical models of galaxy formation predict that feedback from supernovae plays a crucial role in regulating star formation in low-mass galaxies. Efficient feedback, with large outflow rates, is needed in these models to account for the low-efficiency of galaxy formation in low-mass halos. Current constraints on mass outflow rates in this low mass regime are very sparse and the best estimates are uncertain by orders of magnitude. In order to test feedback models, we selected a unique sample of ~50 star-forming galaxies all near background quasars. The advantage in using background quasars over traditional galaxy spectra is that the kinematics and location of the cool gas is known, given by the impact parameter. Hence, the mass outflow rates can be determined more accurately with this technique.Our sample is the largest sample of galaxy-quasar pairs with deep IFU kinematics of the host galaxy and the line-of-sight kinematics with high-resolution optical spectroscopy. With these data we will answer the fundamental question: Are the mass outflow rates higher than the star-formation rates as predicted by theory? More importantly, we will be able to determine whether these depend on the galaxy mass or star-formation rate. Hence, our results will be a huge step forward and provide a strong bench mark for models.Furthermore, the ""Multi Unit Spectroscopic Explorer"" (MUSE) -- to be commissioned in 2013 --with its unprecedented overall throughput, excellent stability and large field of view will be a very powerful tool to study galactic winds. Indeed, it will provide exquisite data on dozens of galaxy-galaxy pairs and quasar-galaxy pairs. Because, the host institute IRAP (Toulouse, FR) is part of the MUSE consortium, and has experts in the field, it will be an ideal place to conduct this research program."