AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Adversarial Learning of Privacy-Preserving and Task-Oriented Representations

Taihong Xiao, Yi-Hsuan Tsai, Kihyuk Sohn, Manmohan Chandraker, Ming-Hsuan Yang

AAAI 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Membership inference and inversion · 73 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add negative pixel reconstruction loss to encoder trainingResNet-50 facial-attribute encoder · CelebA test, 40 attributes; no GAN loss for attackerFace similarity: 0.551 -> 0.515Improved
Add negative pixel reconstruction loss to encoder trainingResNet-50 facial-attribute encoder · CelebA test, 40 attributes; no GAN loss for attackerMean MCC: 0.641 -> 0.612Hurt
Add negative pixel reconstruction loss to encoder trainingResNet-50 facial-attribute encoder · CelebA test, 40 attributes; GAN-trained attackerFace similarity: 0.585 -> 0.513Improved
Add negative pixel reconstruction loss to encoder trainingResNet-50 facial-attribute encoder · CelebA test, SmilingSmiling MCC: 0.851 -> 0.862Improved
Add negative pixel reconstruction loss to encoder trainingResNet-50 facial-attribute encoder · CelebA test, SmilingFace similarity: 0.460 -> 0.424Improved
Provide extra MS-Celeb-1M data to train the decoder attacker (ablation)Adversarial reconstruction defense · CelebA test, 40 attributesFace similarity: 0.551 -> 0.594Hurt
Provide extra MS-Celeb-1M data to train both encoder and decoder (ablation)Adversarial reconstruction defense · CelebA test, 40 attributesFace similarity: 0.541 -> 0.550Improved
Train for one utility attribute rather than 40 attributes (ablation)BCE facial-attribute encoder · CelebA testFace similarity: 0.551 -> 0.460Improved
Increase negative perceptual-loss weight from 0 to 1 (ablation)Adversarial reconstruction defense · CelebA test, 40 attributesFace similarity: 0.631 -> 0.575Improved
Increase negative perceptual-loss weight from 0 to 1 (ablation)Adversarial reconstruction defense · CelebA test, 40 attributesMean MCC: 0.631 -> 0.582Hurt
Increase negative perceptual-loss weight from 1 to 10 (ablation)Adversarial reconstruction defense · CelebA test, 40 attributesFace similarity: 0.575 -> 0.294Improved
Increase negative perceptual-loss weight from 1 to 10 (ablation)Adversarial reconstruction defense · CelebA test, 40 attributesMean MCC: 0.582 -> 0.000Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy