Mechanic: A Learning Rate Tuner

We introduce a technique for tuning the learning rate scale factor of any base optimization algorithm and schedule automatically, which we call $\text{textsc}mechanic$. Our method provides a practical realization of recent theoretical reductions for accomplishing a similar goal in online convex optimization. We rigorously evaluate $\text{textsc}mechanic$ on a range of large scale deep learning tasks with varying batch sizes, schedules, and base optimization algorithms. These experiments demonstrate that depending on the problem, $\text{textsc}mechanic$ either comes very close to, matches or even improves upon manual tuning of learning rates.