Artificial Intelligence

Modern deep neural networks are overparameterized and thus require data augmentation techniques to prevent over-fitting and improve generalization ability. Generative adversarial networks (GANs) are famous for generating visually realistic images. However, the generated images lack diversity and have uncertain class labels. On the other hand, recent methods mix labels proportionally to the salient region.

The ultrasound video data were collected from two sets of neck ultrasound videos of ten healthy subjects at the Ultrasound Department of Longhua Hospital Affiliated to Shanghai University of Traditional Chinese Medicine. Each subject included video files of two groups of LSCM, LSSCap, RSCM, and RSSCap. The video format is avi.

The MRI training data were sourced from three hospitals: Longhua Hospital, Shanghai University of Traditional Chinese Medicine; Huadong Hospital, Fudan University; and Shenzhen Traditional Chinese Medicine Hospital.

This paper develops a correct-by-design controller for an autonomous vehicle interacting with opponent vehicles with unknown intentions. We define an intention-aware control problem incorporating epistemic uncertainties of the opponent vehicles and model their intentions as discrete-valued random variables. Then, we focus on a control objective specified as belief-space temporal logic specifications. From this stochastic control problem, we derive a sound deterministic control problem using stochastic expansion and solve it using shrinking-horizon model predictive control.

In this dataset we release the data of a sequence of boxes that go through a physical binary sorter that acts as a load balancer between warehouses. This particular physical binary sorter works in real-time operating 4.5 million boxes per year. This is a particular example for a company with hundreds of \textit{physical binary sorters} that are central to the internal logistics of the business.