Blockchain is a pillar in 4IR (4th Industrial Revolution) of the digital world. Selection of the right blockchain platform, for anonymity, is a challenge that can be overcome while getting acknowledged how anonymous a...
详细信息
While bias in artificial intelligence is gaining attention across applications, model fairness is especially concerning in medical applications because a person's health may depend on the model outcome. Sources of...
详细信息
In response to the needs of fields such as intersatellite high-speed communication, relay bidirectional high-speed data transmission, ground data transmission, ground measurement and control, and comprehensive informa...
详细信息
Game-Based Learning has already demonstrated its virtues and challenges in Computing education in several disciplines. In this paper, we present a serious game designed to teach computer Network concepts using a 3D vi...
详细信息
With the remarkable advancements in Artificial Intelligence (AI) and Machine Learning (ML) technologies across various fields, these tools offer innovative directions for long-standing, experience-based designs in tun...
详细信息
This paper reviews interdisciplinary collaboration from soft magnetic material engineering to CMOS integrated circuit design level aiming for high power density and high efficiency switching power supplies beyond 10MH...
详细信息
Intelligent learning environments supported by intelligent technology extend the traditional face-to-face dialogues and communication methods, expand the space of intelligent learning, and serve as a bridge to stimula...
详细信息
Visual AI systems are vulnerable to natural and synthetic physical corruption in the real-world. Such corruption often arises unexpectedly and alters the model's performance. In recent years, the primary focus has...
详细信息
ISBN:
(纸本)9781665488679
Visual AI systems are vulnerable to natural and synthetic physical corruption in the real-world. Such corruption often arises unexpectedly and alters the model's performance. In recent years, the primary focus has been on adversarial attacks. However, natural corruptions (e.g., snow, fog, dust) are an omnipresent threat to visual AI systems and should be considered equally important. Many existing works propose interesting solutions to train robust models against natural corruption. These works either leverage image augmentations, which come with the additional cost of model training, or place suspicious patches in the scene to design unadversarial examples. In this work, we propose the idea of naturalistic support artifacts (NSA) for robust prediction. The NSAs are shown to be beneficial in scenarios where model parameters are inaccessible and adding artifacts in the scene is feasible. The NSAs are natural looking objects generated through artifact training using DC-GAN to have high visual fidelity in the scene. We test against natural corruptions on the Imagenette dataset and observe the improvement in prediction confidence score by four times. We also demonstrate NSA's capability to increase adversarial accuracy by 8% on average. Lastly, we qualitatively analyze NSAs using saliency maps to understand how they help improve prediction confidence.
With the development of information technology and the explosive growth of data resources, big data provides unprecedented research perspectives and tools for cognitive psychology. Since traditional research methods i...
详细信息
Systems with automated recommendations are widely used in many mobile applications, such as e-commerce, news, agriculture, and other fields. Agricultural information is an essential driving force for the development o...
详细信息
暂无评论