Rising temperature especially in summer is currently a hot debate. Scientists around the world have raised concerns about Heat Stress Assessment (HSA). It depends on urban geometry, building materials, greenery, environmental factor of region, psychological and behavioral factors inhabitants. Effective accurate heat stress forecasts are useful for managing thermal comfort area. A widely used te...

Abstract Landslides are notoriously difficult to predict because numerous spatially and temporally varying factors contribute slope stability. Artificial neural networks (ANN) have been shown improve prediction accuracy but largely uninterpretable. Here we introduce an additive ANN optimization framework assess landslide susceptibility, as well dataset division outcome interpretation techniques...

Existing neural machine translation (NMT) systems utilize sequence-to-sequence networks to generate target word by word, and then make the generated at each time-step counterpart in references as consistent possible. However, trained model tends focus on ensuring accuracy of current does not consider its future cost which means expected generating subsequent (i.e., next word). To respond this i...

We introduce a new method for speeding up the inference of deep neural networks. It is somewhat inspired by reduced-order modeling techniques dynamical systems. The cornerstone proposed maximum volume algorithm. demonstrate efficiency on networks pre-trained different datasets. show that in many practical cases it possible to replace convolutional layers with much smaller fully-connected relati...