Retrieval-Augmented Convolutional Neural Networks for Improved Robustness against Adversarial Examples
By Jake Zhao, Kyunghyun Cho. 2018
20 citations recorded by the source corpus.
We propose a retrieval-augmented convolutional network and propose to train it with local mixup, a novel variant of the recently proposed mixup algorithm. The proposed hybrid architecture combining a convolutional network and an off-the-shelf retrieval engine was designed to mitigate the adverse effect of off-manifold adversarial examples, while the proposed local mixup addresses on-manifold ones by explicitly encouraging the classifier to locally behave linearly on the data manifold. Our evaluation of the proposed approach against five readily-available adversarial attacks on three datasets--CIFAR-10, SVHN and ImageNet--demonstrate the improved robustness compared to the vanilla convolutional network.
Subfields: Artificial Intelligence, Retrieval-Augmented Generation, Safety, Robustness & Interpretability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Luke S. Zettlemoyer
- Noah A. Smith
- Mohit Bansal
- Michael Lewis
- Percy Liang
- Eric Wallace
- Yejin Choi
- Hannaneh Hajishirzi
In Retrieval-Augmented Generation
In Safety, Robustness & Interpretability
- Eric Wallace
- Jordan L. Boyd-Graber
- Shi Feng
- Dan Jurafsky
- Yoshua Bengio
- Christopher Beckham
- David Lopez-Paz
- Luke S. Zettlemoyer
Citations in the corpus
- Generalization through Memorization: Nearest Neighbor Language Models (2019, methodology): “Similar k NN models to ours have been proposed for computer vision tasks (Papernot & McDaniel, 2018; Orhan, 2018; Zhao & Cho, 2018), primarily motivated by improving interpretability and robustness to adversarial attacks.”
- Manifold Mixup: Better Representations by Interpolating Hidden States (2018, methodology): “However, an important difference is that Manifold Mixup backpropagates gradients through the earlier parts of the network (the layers before the point where mixup is applied), unlike (Zhao & Cho, 2018).”
- A Simple Cache Model for Image Recognition (2018, methodology): “Two recent papers have used cache memories to improve the robustness of image recognition models against adversarial attacks [12, 18].”
- Interpreting Neural Networks with Nearest Neighbors (2018, background): “Retrieval-Augmented Neural Networks (Zhao and Cho, 2018) are similar to DKNN: they augment model predictions with an information retrieval system that searches over network activations from the training data.”