Knowledge distillation (KD) has proved to be an effective approach for deep neural network compression, which learns a compact (student) by transferring the knowledge from pre-trained, over-parameterized (teacher). In traditional KD, transferred is usually obtained feeding training samples teacher obtain class probabilities. However, original dataset not always available due storage costs or pr...

Due to the world’s population growth, excessive solid waste generation is a serious environmental issue. The landfill leachate infiltrates soils, pollutes groundwater, and puts all living things at risk. This study investigates geotechnical properties of soils (marine clay granitic residual soil) migration cadmium (Cd) using high-speed centrifuge column test. All soil samples were subjected phy...

A regional general circulation model (GCM) of the Indian Ocean is used to investigate the influence of prescribed diapycnal diffusivity (Kd) on quasi-steady states of the meridional overturning circulation (MOC). The model has open boundaries at 35°S and 123°E where velocity, temperature, and salinity are prescribed at each time step. The results suggest that quasi-steady overturning states in ...

The Scene Graph Generation (SGG) task aims to detect all the objects and their pairwise visual relationships in a given image. Although SGG has achieved remarkable progress over last few years, almost existing models follow same training paradigm: they treat both object predicate classification as single-label problem, ground-truths are one-hot target labels. However, this prevalent paradigm ov...

Ultraviolet (UV) radiation has a profound impact on marine life, but historically and even currently, most ocean color satellites cannot provide radiance measurements in the UV, thus UV penetration, global ocean. We develop system (termed as UVISRdl) this study, based deep learning, to estimate remote sensing reflectance (Rrs) at 360, 380, 400 nm (collectively termed near-blue bands, nbUV) from...