Adversarial Machine Learning for Deep Hashing

Related Papers of adversarial machine learning in deep hashing based retrieval.

Adversarial Attack

• Adversarial Examples for Hamming Space Search (TCYB 2020)
  This paper provided the first adversarial attack method (non-targeted attack) in deep hashing based retrieval.
• Targeted Attack for Deep Hashing Based Retrieval (ECCV 2020)
  This paper proposed the first targeted attack method for deep hashing based retrieval.
• Prototype-Supervised Adversarial Network for Targeted Attack of Deep Hashing (CVPR 2021)
  The authors proposed a novel prototype-supervised adversarial network (ProS-GAN), which formulates a flexible generative architecture for efficient and effective targeted hashing attack.
• Targeted Attack of Deep Hashing Via Prototype-Supervised Adversarial Networks (TMM 2022)
  This is the extension work of ProS-GAN.
• You See What I Want You To See: Exploring Targeted Black-Box Transferability Attack for Hash-Based Image Retrieval Systems (CVPR 2021)
  This paper started from an adversarial standpoint to explore and enhance the capacity of targeted black-box transferability attack for deep hashing.
• Precise Target-Oriented Attack against Deep Hashing-based Retrieval (MM 2023)
  The authors proposed a novel Precise Target-Oriented Attack dubbed PTA to enhance the precision of targeted attacks with a single target label.
• Targeted Transferable Attack against Deep Hashing Retrieval (MMAsia 2023)
  This paper focused on targeted black-box attack based on transferability and proposed a novel Targeted Transferable Attack method against deep hashing with Generative Adversarial Network (TTAGAN).
• Exploring Targeted Universal Adversarial Attack for Deep Hashing (ICASSP 2024)
  This paper took the first attempt on the more efficient and malicious targeted universal adversarial attack (TUAA) for deep hashing.

Adversarial Defense

• Targeted Attack and Defense for Deep Hashing (SIGIR 2021)
  This paper presented the first adversarial training algorithm to improve the adversarial robustness of deep hashing networks.
• CgAT: Center-Guided Adversarial Training for Deep Hashing-Based Retrieval (WWW 2023)
  The authors presented a min-max based Center-guided Adversarial Training, namely CgAT, to improve the robustness of deep hashing networks through worst adversarial examples.
• Semantic-Aware Adversarial Training for Reliable Deep Hashing Retrieval (TIFS 2023)
  The authors explored Semantic-Aware Adversarial Training (SAAT) for improving the adversarial robustness of deep hashing models.