This paper proposes a learning-based approach for reconstruction of global illumination with very low sampling budgets (as low as 1 spp) at interactive rates. At 1 sample per pixel (spp), the Monte ...

Bryan Lim, Stefan Zohren and Stephen Roberts propose a real-time indicator to detect temporary increases in asset co-movements that they call the autoencoder reconstruction ratio (ARR), which measures ...

Furthermore, other detection methods rely on data augmentation or specialized training techniques which must be asserted before training time. In contrast, we use subset scanning methods from the ...

Serial-autoencoder for personalized recommendation. Higher Education Press . Journal Frontiers of Computer Science DOI 10.1007/s11704-023-2441-1 ...