Greedy Step Averaging: A parameter-free stochastic optimization method
This addresses the need for simpler, more efficient optimization algorithms in machine learning, though it appears incremental as it builds on existing gradient-based methods.
The paper tackles the problem of hyperparameter tuning in stochastic optimization by introducing Greedy Step Averaging (GSA), a parameter-free method that avoids manual learning rate adjustment and shows robustness in experiments on logistic and softmax regression across 16 datasets.
In this paper we present the greedy step averaging(GSA) method, a parameter-free stochastic optimization algorithm for a variety of machine learning problems. As a gradient-based optimization method, GSA makes use of the information from the minimizer of a single sample's loss function, and takes average strategy to calculate reasonable learning rate sequence. While most existing gradient-based algorithms introduce an increasing number of hyper parameters or try to make a trade-off between computational cost and convergence rate, GSA avoids the manual tuning of learning rate and brings in no more hyper parameters or extra cost. We perform exhaustive numerical experiments for logistic and softmax regression to compare our method with the other state of the art ones on 16 datasets. Results show that GSA is robust on various scenarios.