no code implementations • 7 Dec 2023 • Chau-Wai Wong, Chang-Hong Fu, Mengting Xu, Guan-Ming Su
Reshaping, a point operation that alters the characteristics of signals, has been shown capable of improving the compression ratio in video coding practices.
no code implementations • 19 Nov 2022 • Zhongnian Li, Jian Zhang, Mengting Xu, Xinzheng Xu, Daoqiang Zhang
In this paper, we propose a novel problem setting called Complementary Labels Learning with Augmented Classes (CLLAC), which brings the challenge that classifiers trained by complementary labels should not only be able to classify the instances from observed classes accurately, but also recognize the instance from the Augmented Classes in the testing phase.
no code implementations • 23 Jun 2022 • Mengting Xu, Tao Zhang, Zhongnian Li, Daoqiang Zhang
Therefore, guaranteeing the robustness of hard examples is crucial for improving the final robustness of the model.
no code implementations • 29 Jan 2022 • Mengting Xu, Tao Zhang, Zhongnian Li, Daoqiang Zhang
Further, we propose Scale-Invariant (SI) adversarial defense mechanism based on the cosine angle matrix, which can be embedded into the popular adversarial defenses.
no code implementations • 29 Nov 2021 • Mengting Xu, Tao Zhang, Daoqiang Zhang
However, the defense methods that have good effect in natural images may not be suitable for medical diagnostic tasks.
1 code implementation • 5 Mar 2021 • Mengting Xu, Tao Zhang, Zhongnian Li, Mingxia Liu, Daoqiang Zhang
Deep learning models (with neural networks) have been widely used in challenging tasks such as computer-aided disease diagnosis based on medical images.
no code implementations • 24 Dec 2020 • Mengting Xu, Tao Zhang, Zhongnian Li, Daoqiang Zhang
A range of defense methods have been proposed to improve the robustness of neural networks on adversarial examples, among which provable defense methods have been demonstrated to be effective to train neural networks that are certifiably robust to the attacker.
no code implementations • 18 Jun 2019 • Hui Li, Mengting Xu, Sourav S. Bhowmick, Changsheng Sun, Zhongyuan Jiang, Jiangtao Cui
As the number of required samples have been recently proven to be lower bounded by a particular threshold that presets tradeoff between the accuracy and efficiency, the result quality of these traditional solutions is hard to be further improved without sacrificing efficiency.