Zobrazeno 1 - 10
of 1 780
pro vyhledávání: '"Sabanayagam A"'
Representation learning aims to extract meaningful lower-dimensional embeddings from data, known as representations. Despite its widespread application, there is no established definition of a ``good'' representation. Typically, the representation qu
Externí odkaz:
http://arxiv.org/abs/2412.03471
Machine learning models are highly vulnerable to label flipping, i.e., the adversarial modification (poisoning) of training labels to compromise performance. Thus, deriving robustness certificates is important to guarantee that test predictions remai
Externí odkaz:
http://arxiv.org/abs/2412.00537
Generalization of machine learning models can be severely compromised by data poisoning, where adversarial changes are applied to the training data. This vulnerability has led to interest in certifying (i.e., proving) that such changes up to a certai
Externí odkaz:
http://arxiv.org/abs/2407.10867
Autor:
Xinyu Zhao, Xingwang Gu, Lihui Meng, Yongwei Chen, Qing Zhao, Shiyu Cheng, Wenfei Zhang, Tiantian Cheng, Chuting Wang, Zhengming Shi, Shengyin Jiao, Changlong Jiang, Guofang Jiao, Da Teng, Xiaolei Sun, Bilei Zhang, Yakun Li, Huiqin Lu, Changzheng Chen, Hao Zhang, Ling Yuan, Chang Su, Han Zhang, Song Xia, Anyi Liang, Mengda Li, Dan Zhu, Meirong Xue, Dawei Sun, Qiuming Li, Ziwu Zhang, Donglei Zhang, Hongbin Lv, Rishet Ahmat, Zilong Wang, Charumathi Sabanayagam, Xiaowei Ding, Tien Yin Wong, Youxin Chen
Publikováno v:
npj Digital Medicine, Vol 7, Iss 1, Pp 1-11 (2024)
Abstract To address challenges in screening for chronic kidney disease (CKD), we devised a deep learning-based CKD screening model named UWF-CKDS. It utilizes ultra-wide-field (UWF) fundus images to predict the presence of CKD. We validated the model
Externí odkaz:
https://doaj.org/article/aca8144e5b2c41ebaf25fc9c7a2baa9d
Test-time defenses are used to improve the robustness of deep neural networks to adversarial examples during inference. However, existing methods either require an additional trained classifier to detect and correct the adversarial samples, or perfor
Externí odkaz:
http://arxiv.org/abs/2307.11672
Autor:
Cagnetta, Francesco, Oliveira, Deborah, Sabanayagam, Mahalakshmi, Tsilivis, Nikolaos, Kempe, Julia
Lecture notes from the course given by Professor Julia Kempe at the summer school "Statistical physics of Machine Learning" in Les Houches. The notes discuss the so-called NTK approach to problems in machine learning, which consists of gaining an und
Externí odkaz:
http://arxiv.org/abs/2307.02693
Understanding the properties of well-generalizing minima is at the heart of deep learning research. On the one hand, the generalization of neural networks has been connected to the decision boundary complexity, which is hard to study in the high-dime
Externí odkaz:
http://arxiv.org/abs/2306.07104
The central question in representation learning is what constitutes a good or meaningful representation. In this work we argue that if we consider data with inherent cluster structures, where clusters can be characterized through different means and
Externí odkaz:
http://arxiv.org/abs/2212.01046
The fundamental principle of Graph Neural Networks (GNNs) is to exploit the structural information of the data by aggregating the neighboring nodes using a `graph convolution' in conjunction with a suitable choice for the network architecture, such a
Externí odkaz:
http://arxiv.org/abs/2210.09809
Autor:
Preeti Gupta, Aurora Chan, Vu Tai-Anh, Ryan E. K. Man, Eva K. Fenwick, Amudha Aravindhan, Chay Junxing, Joanne M. Wood, Alex A. Black, Jia Hui Ng, Ching-Yu Cheng, Charumathi Sabanayagam, Ecosse L. Lamoureux
Publikováno v:
BMC Public Health, Vol 24, Iss 1, Pp 1-12 (2024)
Abstract Background To determine the prevalence, risk factors; and impact on patient health and economic outcomes across the laterality spectrum of multiple sensory impairment (MSI) in a multi-ethnic older Asian population. Methods In this population
Externí odkaz:
https://doaj.org/article/8b699f75523e427fbf7afc3b4cb8b243