Adversarial Meta-Learning

Abstract

Meta-learning enables a model to learn from very limited data to undertake a new task. In this paper, we study the general meta-learning with adversarial samples. We present a meta-learning algorithm, ADML (ADversarial Meta-Learner), which leverages clean and adversarial samples to optimize the initialization of a learning model in an adversarial manner. ADML leads to the following desirable properties: 1) it turns out to be very effective even in the cases with only clean samples; 2) it is robust to adversarial samples, i.e., unlike other meta-learning algorithms, it only leads to a minor performance degradation when there are adversarial samples; 3) it sheds light on tackling the cases with limited and even contaminated samples. It has been shown by extensive experimental results that ADML consistently outperforms three representative meta-learning algorithms in the cases involving adversarial samples, on two widely-used image datasets, MiniImageNet and CIFAR100, in terms of both accuracy and robustness.

arXiv comment11 pages

References in corpus (8)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Theoretical Models of Learning to Learn
• A Simple Neural Attentive Meta-Learner
• Learning with a Strong Adversary
• Meta-Learning and Universality: Deep Representations and Gradient Descent can Approximate any Learning Algorithm
• Are Accuracy and Robustness Correlated?
• Meta Continual Learning

Cited by in corpus (9)

• Adversarially Robust Few-Shot Learning: A Meta-Learning Approach
• Task-Robust Model-Agnostic Meta-Learning
• Adv-Makeup: A New Imperceptible and Transferable Attack on Face Recognition
• ARCADe: A Rapid Continual Anomaly Detector
• Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning
• Adaptive Adversarial Training for Meta Reinforcement Learning
• Adversarial Monte Carlo Meta-Learning of Optimal Prediction Procedures
• Long-term Cross Adversarial Training: A Robust Meta-learning Method for Few-shot Classification Tasks
• On sensitivity of meta-learning to support data