Online Learning of a Memory for Learning Rates

TL;DR

This paper presents an online meta-learning algorithm that creates and updates a memory of optimal learning rates, enabling faster learning across tasks by predicting gradient scaling, applicable in various optimization scenarios.

Contribution

Introduces a computationally efficient online meta-learning method that learns and updates a memory of learning rates to accelerate task-specific learning.

Findings

Speeds up MNIST classification learning.

Improves learning control tasks in batch and online settings.

Can be combined with any gradient-based optimizer.

Abstract

The promise of learning to learn for robotics rests on the hope that by extracting some information about the learning process itself we can speed up subsequent similar learning tasks. Here, we introduce a computationally efficient online meta-learning algorithm that builds and optimizes a memory model of the optimal learning rate landscape from previously observed gradient behaviors. While performing task specific optimization, this memory of learning rates predicts how to scale currently observed gradients. After applying the gradient scaling our meta-learner updates its internal memory based on the observed effect its prediction had. Our meta-learner can be combined with any gradient-based optimizer, learns on the fly and can be transferred to new optimization tasks. In our evaluations we show that our meta-learning algorithm speeds up learning of MNIST classification and a variety of learning control tasks, either in batch or online learning settings.