Leonardo-lyh / AdversarialQuerying

A PyTorch implementation of the method found in "Adversarially Robust Few-Shot Learning: A Meta-Learning Approach"

Adversarially Robust Few-Shot Learning: A Meta-Learning Approach

Micah Goldblum, Liam Fowl, Tom Goldstein

This repository contains PyTorch code for adversarial querying with ProtoNet, R2-D2, and MetaOptNet. Adversarial querying is an algorithm for producing robust meta-learners. More can be found in our NeurIPS 2020 paper. We adapt models and data loading from here.

Prerequisites:

• Python2
• PyTorch
• CUDA