Computer simulations at the fixed point using an inverse renormalization group transformation

Following Brandt and Ron's suggestion of inverting the renormalization group transformation used in Monte Carlo renormalization, it is shown that efficient computer simulations of the fixed point of the transformation can be carried out on very large systems without critical slowing down. We illustrate the new method with calculations of critical exponents for the two- and three-dimensional Ising models, based on several different transformations.