Over the past decade, machine learning has become a cornerstone of innovation in medical imaging. Our lab has been at the forefront of introducing machine learning techniques to the field of X-ray tomographic image reconstruction, as demonstrated through a series of pioneering publications. In 2011, we developed the first dictionary learning-based method for low-dose CT reconstruction, which laid ...

内容简介：As platforms increasingly deploy AI chatbots for transaction negotiations, how users respond to detecting algorithmic counterparts—rather than human agents—remains poorly understood. Analyzing driver call-in data from a freight platform where drivers independently recognized AI identity during price negotiations, this paper examines how drivers’ detection of AI identity affects deal-clo...

Animated virtual reality (VR) stories, combining the presence of VR and the artistry of computer animation, offer a compelling way to deliver messages and evoke emotions. This talk explores how computational creativity support can reduce the high barriers to creating VR stories. I will discuss where, whether, and how computational approaches originally developed for desktop content can be adapted ...

Heterogeneous catalysis is essential to modern chemical engineering, chemistry, andenergy applications. While reaction mechanisms have been extensively studied underultra-high vacuum (UHV)on clean, well-defined surfaces, applying these insights underpractical conditions remains challenging. This is due to (1)the dynamic evolution ofcomplex catalyst surfaces and (2) the difficulty in detecting high...

The presence of data shift between training and test datasets, coupled with model misspecification, can lead to instability in regression predictions across diverse datasets. In this talk, we present a novel subsampling algorithm for stable prediction, which employs uniform design and confounder balancing methods. Theoretic analyses show that the uniform measure minimizes the maximum integral mean...