Zobrazeno 1 - 10
of 270
pro vyhledávání: '"Lee, Hwee Kuan"'
In-Context Learning (ICL) has been a powerful emergent property of large language models that has attracted increasing attention in recent years. In contrast to regular gradient-based learning, ICL is highly interpretable and does not require paramet
Externí odkaz:
http://arxiv.org/abs/2406.02847
Autor:
Coppola, Davide, Lee, Hwee Kuan
This study explores the impact of adversarial perturbations on Convolutional Neural Networks (CNNs) with the aim of enhancing the understanding of their underlying mechanisms. Despite numerous defense methods proposed in the literature, there is stil
Externí odkaz:
http://arxiv.org/abs/2405.20672
Autor:
Zhang, Tianyi, Lyu, Shangqing, Lei, Yanli, Chen, Sicheng, Ying, Nan, He, Yufang, Zhao, Yu, Feng, Yunlu, Lee, Hwee Kuan, Zhang, Guanglei
Pathological image analysis is a crucial field in computer vision. Due to the annotation scarcity in the pathological field, pre-training with self-supervised learning (SSL) is widely applied to learn on unlabeled images. However, the current SSL-bas
Externí odkaz:
http://arxiv.org/abs/2311.06712
Given a poorly documented neural network model, we take the perspective of a forensic investigator who wants to find out the model's data domain (e.g. whether on face images or traffic signs). Although existing methods such as membership inference an
Externí odkaz:
http://arxiv.org/abs/2305.05869
Autor:
Chen, Zerui, Teo, Sonia Xhyn, Ochtman, Andrie, Saw, Shier Nee, Cheng, Nicholas, Lim, Eric Tien Siang, Lyu, Murphy, Lee, Hwee Kuan
A key technology enabling the success of catheter ablation treatment for atrial tachycardia is activation mapping, which relies on manual local activation time (LAT) annotation of all acquired intracardiac electrogram (EGM) signals. This is a time-co
Externí odkaz:
http://arxiv.org/abs/2206.07515
Autor:
Paknezhad, Mahsa, Rengarajan, Hamsawardhini, Yuan, Chenghao, Suresh, Sujanya, Gupta, Manas, Ramasamy, Savitha, Lee, Hwee Kuan
This paper takes a parallel learning approach for robust and transparent AI. A deep neural network is trained in parallel on multiple tasks, where each task is trained only on a subset of the network resources. Each subset consists of network segment
Externí odkaz:
http://arxiv.org/abs/2201.09534
We propose a machine learning method to solve Schrodinger equations for a Hamiltonian that consists of an unperturbed Hamiltonian and a perturbation. We focus on the cases where the unperturbed Hamiltonian can be solved analytically or solved numeric
Externí odkaz:
http://arxiv.org/abs/2106.12094
Publikováno v:
Computerized Medical Imaging and Graphics, 87, 101838 (2021)
Researchers working on computational analysis of Whole Slide Images (WSIs) in histopathology have primarily resorted to patch-based modelling due to large resolution of each WSI. The large resolution makes WSIs infeasible to be fed directly into the
Externí odkaz:
http://arxiv.org/abs/2106.02864
Publikováno v:
Sci. Rep. 11, 9617 (2021)
The inverse renormalization group is studied based on the image super-resolution using the deep convolutional neural networks. We consider the improved correlation configuration instead of spin configuration for the spin models, such as the two-dimen
Externí odkaz:
http://arxiv.org/abs/2104.04482
Autor:
Paknezhad, Mahsa, Ngo, Cuong Phuc, Winarto, Amadeus Aristo, Cheong, Alistair, Beh, Chuen Yang, Wu, Jiayang, Lee, Hwee Kuan
Publikováno v:
Neurocomputing, 2022
Despite many proposed algorithms to provide robustness to deep learning (DL) models, DL models remain susceptible to adversarial attacks. We hypothesize that the adversarial vulnerability of DL models stems from two factors. The first factor is data
Externí odkaz:
http://arxiv.org/abs/2103.00778