On Adversarial Mixup Resynthesis

2019/03/07 by Christopher Beckham, Sina Honari, Beckham, Christopher +14 · 1 citation
Computer Science · Mathematics · #Adversarial Robustness in Machine Learning #Anomaly Detection Techniques and Applications #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML) #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.1903.02709

'Camera-ready draft'

openalex publication_date 2019/03/07 · arxiv created 2019/10/23 · arxiv updated 2019/10/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

In this paper, we explore new approaches to combining information encoded within the learned representations of auto-encoders. We explore models that are capable of combining the attributes of multiple inputs such that a resynthesised output is trained to fool an adversarial discriminator for real versus synthesised data. Furthermore, we explore the use of such an architecture in the context of semi-supervised learning, where we learn a mixing function whose objective is to produce interpolations of hidden states, or masked combinations of latent representations that are consistent with a conditioned class label. We show quantitative and qualitative evidence that such a formulation is an interesting avenue of research.

Cited by

• Inclusive GAN: Improving Data and Minority Coverage in Generative Models

Related

• Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attacks
• Denoising Adversarial Autoencoders
• Decoupled Adversarial Contrastive Learning for Self-supervised Adversarial Robustness
• mixup: Beyond Empirical Risk Minimization
• MixACM: Mixup-Based Robustness Transfer via Distillation of Activated Channel Maps