MearcAll papers

arXiv 1412.6572 · stat.ML cs.LG

Explaining and Harnessing Adversarial Examples

Ian J. Goodfellow, Jonathon Shlens, Christian Szegedy

Claims were extracted automatically by reducer-1 on 2026-11-12. The paper's license does not allow republishing it, so long claims are shortened to a quotation. Read the full text on arXiv.

No reproduction recordedclm_f5f2464b
We argue instead that the primary cause of neural networks’ vulnerability to adversarial perturbation is their linear nature.

ABSTRACT · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_fea7a52e
We show that these speculative hypotheses are unnecessary.

I NTRODUCTION · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_dc0badaf
We show that adversarial training can provide an additional regularization benefit beyond that provided by using dropout (Srivastava et al., 2014) alone.

I NTRODUCTION · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_56fb76d6
Its inference procedure gets good classification accuracy (an 0.88% error rate) on MNIST.

A LTERNATIVE HYPOTHESES · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_ab379097
To solve the problem introduced by Nguyen et al. (2014) of generating a fooling image for a particular class, we propose adding ϵ∇xp(y = i | x) to a Gaussian … (excerpt)

R UBBISH CLASS EXAMPLES · page 11 · Automated extraction, not reviewed · 0 reviewed reproductions

Reproduced one of these claims?

Record your result in the Mearc app on your own machine, then use Publish this result with this paper's id (1412.6572) and the claim id shown next to the claim. Mearc creates a file that holds only what you choose to publish. Add it to the public claim data by pull request; a maintainer other than you reviews it before it counts.