LOSSGRAD: automatic learning rate in gradient descent
This addresses the challenge of manual learning rate tuning for practitioners in machine learning, though it appears incremental as it builds on existing gradient descent methods.
The paper tackles the problem of automatically adjusting the learning rate during gradient descent in neural network training, proposing the LOSSGRAD algorithm that finds locally optimal step-sizes using quadratic approximation, and experimentally shows it is insensitive to initial learning rates while achieving comparable results to other methods.
In this paper, we propose a simple, fast and easy to implement algorithm LOSSGRAD (locally optimal step-size in gradient descent), which automatically modifies the step-size in gradient descent during neural networks training. Given a function $f$, a point $x$, and the gradient $\nabla_x f$ of $f$, we aim to find the step-size $h$ which is (locally) optimal, i.e. satisfies: $$ h=arg\,min_{t \geq 0} f(x-t \nabla_x f). $$ Making use of quadratic approximation, we show that the algorithm satisfies the above assumption. We experimentally show that our method is insensitive to the choice of initial learning rate while achieving results comparable to other methods.