Fetching the paper…

Learning-rate-free Momentum SGD with Reshuffling Converges in Nonsmooth Nonconvex Optimization · Around