Thürmer Andrea

TL;DR: The Unsupervised Descriptor Selection (UDS) is proposed to tackle few-shot learning tasks, and achieves the comparable performance to state-of-the-art methods, and improves the performance of prior meta-learning methods.

TL;DR: In this article , a descriptor selection module is proposed to localize and select semantic meaningful regions in feature maps without supervision, and the selected features are then mapped into novel vectors by a task-related aggregation module to enhance internal representations.

TL;DR: In this article, a full coupled three-dimensional model incorporating with buoyancy flow was established to obtain the actual airflow and temperature field of high-temperature tunnel, and the authors highlighted the significant effect of the presence and magnitude of groundwater flow on temperature variations and heat transfer of tunnel.

TL;DR: In this paper , a full coupled three-dimensional model incorporating with buoyancy flow was established to obtain the actual airflow and temperature field of high-temperature tunnel, and an orthogonal test was performed to investigate the influence degree of major factors on heat exchange of tunnel.

TL;DR: Saliva samples obtained from 178 patients with ESCC and 101 healthy controls were allocated to screening and verification cohorts, and Q-PCR assay showed significantly higher numbers of Streptococcus salivarius, Fusobacterium nucleatum and Porphyromonas gingivalis in patients withESCC, when compared with healthy controls.