Task-Robust Model-Agnostic Meta-Learning
This work addresses the robustness issue in meta-learning for scenarios where task distributions may shift between training and testing, which is incremental as it builds on existing MAML methods.
The paper tackles the problem of meta-learning methods being sensitive to task distribution shifts by introducing a task-robust version of Model-Agnostic Meta-Learning (MAML) that minimizes the maximum loss over meta-training tasks, resulting in improved worst-case performance and robustness to distribution shifts, as demonstrated in experiments with sinusoid regression and image classification.
Meta-learning methods have shown an impressive ability to train models that rapidly learn new tasks. However, these methods only aim to perform well in expectation over tasks coming from some particular distribution that is typically equivalent across meta-training and meta-testing, rather than considering worst-case task performance. In this work we introduce the notion of "task-robustness" by reformulating the popular Model-Agnostic Meta-Learning (MAML) objective [Finn et al. 2017] such that the goal is to minimize the maximum loss over the observed meta-training tasks. The solution to this novel formulation is task-robust in the sense that it places equal importance on even the most difficult and/or rare tasks. This also means that it performs well over all distributions of the observed tasks, making it robust to shifts in the task distribution between meta-training and meta-testing. We present an algorithm to solve the proposed min-max problem, and show that it converges to an $ε$-accurate point at the optimal rate of $\mathcal{O}(1/ε^2)$ in the convex setting and to an $(ε, δ)$-stationary point at the rate of $\mathcal{O}(\max\{1/ε^5, 1/δ^5\})$ in nonconvex settings. We also provide an upper bound on the new task generalization error that captures the advantage of minimizing the worst-case task loss, and demonstrate this advantage in sinusoid regression and image classification experiments.