Search Results for author:
Found 8 papers, 4 papers with code
MEA-Defender: A Robust Watermark against Model Extraction Attack
To protect the Intellectual Property (IP) of the original owners over such DNN models, backdoor-based watermarks have been extensively studied.
DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models
Dataset sanitization is a widely adopted proactive defense against poisoning-based backdoor attacks, aimed at filtering out and removing poisoned samples from training datasets.
A Novel Membership Inference Attack against Dynamic Neural Networks by Utilizing Policy Networks Information
Although traditional static DNNs are vulnerable to the membership inference attack (MIA) , which aims to infer whether a particular point was used to train the model, little is known about how such an attack performs on the dynamic NNs.
SSL-WM: A Black-Box Watermarking Approach for Encoders Pre-trained by Self-supervised Learning
Recent years have witnessed tremendous success in Self-Supervised Learning (SSL), which has been widely utilized to facilitate various downstream tasks in Computer Vision (CV) and Natural Language Processing (NLP) domains.
Invisible Backdoor Attacks Using Data Poisoning in the Frequency Domain
However, most of the triggers used in the current study are fixed patterns patched on a small fraction of an image and are often clearly mislabeled, which is easily detected by humans or defense methods such as Neural Cleanse and SentiNet.
DBIA: Data-free Backdoor Injection Attack against Transformer Networks
In this paper, we propose DBIA, a novel data-free backdoor attack against the CV-oriented transformer networks, leveraging the inherent attention mechanism of transformers to generate triggers and injecting the backdoor using the poisoned surrogate dataset.
HufuNet: Embedding the Left Piece as Watermark and Keeping the Right Piece for Ownership Verification in Deep Neural Networks
Most existing solutions embed backdoors in DNN model training such that DNN ownership can be verified by triggering distinguishable model behaviors with a set of secret inputs.
Efficient Computation of Quantized Neural Networks by {−1, +1} Encoding Decomposition
Deep neural networks require extensive computing resources, and can not be efficiently applied to embedded devices such as mobile phones, which seriously limits their applicability.
