On the Robustness of ML-Based Network Intrusion Detection Systems: An Adversarial and Distribution Shift Perspective

M Wang, N Yang, DH Gunasinghe, N Weng - Computers, 2023 - mdpi.com
Utilizing machine learning (ML)-based approaches for network intrusion detection systems
(NIDSs) raises valid concerns due to the inherent susceptibility of current ML models to …

Causal-structure driven augmentations for text ood generalization

A Feder, Y Wald, C Shi, S Saria… - Advances in Neural …, 2024 - proceedings.neurips.cc
The reliance of text classifiers on spurious correlations can lead to poor generalization at
deployment, raising concerns about their use in safety-critical domains such as healthcare …

Domain generalization for medical image analysis: A survey

JS Yoon, K Oh, Y Shin, MA Mazurowski… - arXiv preprint arXiv …, 2023 - arxiv.org
Medical Image Analysis (MedIA) has become an essential tool in medicine and healthcare,
aiding in disease diagnosis, prognosis, and treatment planning, and recent successes in …

Causality and independence enhancement for biased node classification

G Chen, Y Wang, F Guo, Q Guo, J Shao… - Proceedings of the …, 2023 - dl.acm.org
Most existing methods that address out-of-distribution (OOD) generalization for node
classification on graphs primarily focus on a specific type of data biases, such as label …

Enhancing minority classes by mixing: an adaptative optimal transport approach for long-tailed classification

J Gao, H Zhao, Z Li, D Guo - Advances in Neural …, 2024 - proceedings.neurips.cc
Real-world data usually confronts severe class-imbalance problems, where several majority
classes have a significantly larger presence in the training set than minority classes. One …

Domain constraints improve risk prediction when outcome data is missing

S Balachandar, N Garg, E Pierson - arXiv preprint arXiv:2312.03878, 2023 - arxiv.org
Machine learning models are often trained to predict the outcome resulting from a human
decision. For example, if a doctor decides to test a patient for disease, will the patient test …

Channel Vision Transformers: An Image Is Worth C x 16 x 16 Words

Y Bao, S Sivanandan, T Karaletsos - arXiv preprint arXiv:2309.16108, 2023 - arxiv.org
Vision Transformer (ViT) has emerged as a powerful architecture in the realm of modern
computer vision. However, its application in certain imaging fields, such as microscopy and …

Nuisances via negativa: Adjusting for spurious correlations via data augmentation

A Puli, N Joshi, Y Wald, H He, R Ranganath - arXiv preprint arXiv …, 2022 - arxiv.org
In prediction tasks, there exist features that are related to the label in the same way across
different settings for that task; these are semantic features or semantics. Features with …

" why did the model fail?": Attributing model performance changes to distribution shifts

H Zhang, H Singh, M Ghassemi, S Joshi - 2023 - proceedings.mlr.press
Abstract Machine learning models frequently experience performance drops under
distribution shifts. The underlying cause of such shifts may be multiple simultaneous factors …

Reliable Model Watermarking: Defending Against Theft without Compromising on Evasion

H Zhu, S Liang, W Hu, L Fangqi, J Jia… - Proceedings of the 32nd …, 2024 - dl.acm.org
With the rise of Machine Learning as a Service (MLaaS) platforms, safeguarding the
intellectual property of deep learning models is becoming paramount. Among various …