Devil is in channels: Contrastive single domain generalization for medical image segmentation
Deep learning-based medical image segmentation models suffer from performance
degradation when deployed to a new healthcare center. To address this issue …
degradation when deployed to a new healthcare center. To address this issue …
Mitigating adversarial attacks in federated learning with trusted execution environments
S Queyrut, V Schiavoni, P Felber - 2023 IEEE 43rd …, 2023 - ieeexplore.ieee.org
The main premise of federated learning (FL) is that machine learning model updates are
computed locally to preserve user data privacy. This approach avoids by design user data to …
computed locally to preserve user data privacy. This approach avoids by design user data to …
Bridging the gap: Rademacher complexity in robust and standard generalization
Abstract Training Deep Neural Networks (DNNs) with adversarial examples often results in
poor generalization to test-time adversarial data. This paper investigates this issue, known …
poor generalization to test-time adversarial data. This paper investigates this issue, known …
Phase-aware adversarial defense for improving adversarial robustness
Deep neural networks have been found to be vulnerable to adversarial noise. Recent works
show that exploring the impact of adversarial noise on intrinsic components of data can help …
show that exploring the impact of adversarial noise on intrinsic components of data can help …
Improving the robustness of transformer-based large language models with dynamic attention
Transformer-based models, such as BERT and GPT, have been widely adopted in natural
language processing (NLP) due to their exceptional performance. However, recent studies …
language processing (NLP) due to their exceptional performance. However, recent studies …
Theoretical Analysis of Robust Overfitting for Wide DNNs: An NTK Approach
Adversarial training (AT) is a canonical method for enhancing the robustness of deep neural
networks (DNNs). However, recent studies empirically demonstrated that it suffers from …
networks (DNNs). However, recent studies empirically demonstrated that it suffers from …
Specification overfitting in artificial intelligence
Abstract Machine learning (ML) and artificial intelligence (AI) approaches are often criticized
for their inherent bias and for their lack of control, accountability, and transparency …
for their inherent bias and for their lack of control, accountability, and transparency …
How robust accuracy suffers from certified training with convex relaxations
P De Bartolomeis, J Clarysse, A Sanyal… - arXiv preprint arXiv …, 2023 - arxiv.org
Adversarial attacks pose significant threats to deploying state-of-the-art classifiers in safety-
critical applications. Two classes of methods have emerged to address this issue: empirical …
critical applications. Two classes of methods have emerged to address this issue: empirical …
[PDF][PDF] Harmonic Analysis With Neural Semi-CRF.
Automatic harmonic analysis of symbolic music is an important and useful task for both
composers and listeners. The task consists of two components: recognizing harmony labels …
composers and listeners. The task consists of two components: recognizing harmony labels …
Pelta: shielding transformers to mitigate evasion attacks in federated learning
S Queyrut, YD Bromberg, V Schiavoni - Proceedings of the 3rd …, 2022 - dl.acm.org
The main premise of federated learning is that machine learning model updates are
computed locally, in particular to preserve user data privacy, as those never leave the …
computed locally, in particular to preserve user data privacy, as those never leave the …