Task-Robust Model-Agnostic Meta-Learning

Abstract

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 $\epsilon$-accurate point at the optimal rate of $\mathcal{O}(1/{\epsilon }^2)$ in the convex setting and to an $(\epsilon ,\delta )$-stationary point at the rate of $\mathcal{O}(\max \{1/{\epsilon }^5,1/{\delta }^5\})$ 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.

References in corpus (16)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Theoretical Models of Learning to Learn
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• Learning to reinforcement learn
• Adversarial Attacks on Graph Neural Networks via Meta Learning
• Agnostic Federated Learning
• Meta-Learning with Implicit Gradients
• Projection onto the probability simplex: An efficient algorithm with a simple proof, and an application
• Solving a Class of Non-Convex Min-Max Games Using Iterative First Order Methods
• Online Meta-Learning
• ES-MAML: Simple Hessian-Free Meta Learning
• On the Convergence Theory of Gradient-Based Model-Agnostic Meta-Learning Algorithms
• Minimizing the Maximal Loss: How and Why?
• Adversarial Meta-Learning
• Weighted Meta-Learning

Cited by in corpus (9)

• Theoretical Convergence of Multi-Step Model-Agnostic Meta-Learning
• Is Support Set Diversity Necessary for Meta-Learning?
• How Fine-Tuning Allows for Effective Meta-Learning
• Generalization of Model-Agnostic Meta-Learning Algorithms: Recurring and Unseen Tasks
• How Does the Task Landscape Affect MAML Performance?
• Improved Robustness and Safety for Pre-Adaptation of Meta Reinforcement Learning with Prior Regularization
• Towards Enabling Meta-Learning from Target Models
• Compositional federated learning: Applications in distributionally robust averaging and meta learning
• Transfer Bayesian Meta-learning via Weighted Free Energy Minimization