We investigate the design aspects of feature distillation methods achieving network compression and propose a novel feature distillation method in which the distillation loss is designed to make a synergy among various aspects: teacher transform, student transform, distillation feature position and distance function. Our proposed distillation loss includes a feature transform with a newly designed margin ReLU, a new distillation feature position, and a partial L 2 distance function to skip redundant information giving adverse effects to the compression of student. In Ima-geNet, our proposed method achieves 21.65% of top-1 error with ResNet50, which outperforms the performance of the teacher network, ResNet152. Our proposed method is evaluated on various tasks such as image classification, object detection and semantic segmentation and achieves a significant performance improvement in all tasks. The code is available at bhheo.github.io/overhaul * This work was done when authors were in research internship at Clova AI Research, NAVER corp. 78.31 Teacher (ResNet152) 1 arXiv:1904.01866v2 [cs.CV] 9 Aug 2019 T t T s T t T s d d Training image Y t Y s T t T s d Teacher network Student network Figure 2. The general training scheme of feature distillation. The form of teacher transform Tt, student transform Ts and distance d differ from method to method.

Highlights d We identify viral and host proteins that directly interact with coronavirus RNAs d Comparison of SARS-CoV-2 and HCoV-OC43 shows conservation of coronavirus RNA interactome d This reveals 17 antiviral factors such as LARP1, ZC3HAV1, TRIM25, PARP12, and SHFL d We also uncover 9 proviral factors hijacked by SARS-CoV-2, including EIF3D and CSDE1 Authors Sungyul Lee, Young-suk Lee,

No abstract