Pith. sign in

REVIEW

Deep Adversarially-Enhanced k-Nearest Neighbors

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2108.06797 v2 pith:UOWSJ7SW submitted 2021-08-15 cs.LG cs.AIcs.CV

Deep Adversarially-Enhanced k-Nearest Neighbors

classification cs.LG cs.AIcs.CV
keywords deepdaeknnk-nearestneighborsrobustness-accuracytrade-offadversarially-enhanceddknn
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Recent works have theoretically and empirically shown that deep neural networks (DNNs) have an inherent vulnerability to small perturbations. Applying the Deep k-Nearest Neighbors (DkNN) classifier, we observe a dramatically increasing robustness-accuracy trade-off as the layer goes deeper. In this work, we propose a Deep Adversarially-Enhanced k-Nearest Neighbors (DAEkNN) method which achieves higher robustness than DkNN and mitigates the robustness-accuracy trade-off in deep layers through two key elements. First, DAEkNN is based on an adversarially trained model. Second, DAEkNN makes predictions by leveraging a weighted combination of benign and adversarial training data. Empirically, we find that DAEkNN improves both the robustness and the robustness-accuracy trade-off on MNIST and CIFAR-10 datasets.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.