Bootstrapped Meta-Learning

TL;DR

This paper introduces a bootstrapped meta-learning algorithm that improves efficiency and performance by enabling the meta-learner to teach itself, extending meta-learning horizons and enhancing exploration strategies.

Contribution

It proposes a novel bootstrapping approach for meta-learning that guarantees performance improvements and reduces computational complexity, advancing state-of-the-art results.

Findings

Achieved new state-of-the-art on Atari ALE benchmark.

Demonstrated efficiency gains in multi-task meta-learning.

Enabled meta-learned exploration in Q-learning without backpropagation.

Abstract

Meta-learning empowers artificial intelligence to increase its efficiency by learning how to learn. Unlocking this potential involves overcoming a challenging meta-optimisation problem. We propose an algorithm that tackles this problem by letting the meta-learner teach itself. The algorithm first bootstraps a target from the meta-learner, then optimises the meta-learner by minimising the distance to that target under a chosen (pseudo-)metric. Focusing on meta-learning with gradients, we establish conditions that guarantee performance improvements and show that the metric can control meta-optimisation. Meanwhile, the bootstrapping mechanism can extend the effective meta-learning horizon without requiring backpropagation through all updates. We achieve a new state-of-the art for model-free agents on the Atari ALE benchmark and demonstrate that it yields both performance and efficiency gains in multi-task meta-learning. Finally, we explore how bootstrapping opens up new possibilities and find that it can meta-learn efficient exploration in an epsilon-greedy Q-learning agent, without backpropagating through the update rule.