The paper introduces TimeHUT, a new method for learning time-series representations using hierarchical uniformity-tolerance balancing of contrastive representations. TimeHUT employs a hierarchical setup to learn both instance-wise and temporal information, along with a temperature scheduler to balance uniformity and tolerance. The method was evaluated on UCR, UAE, Yahoo, and KPI datasets, demonstrating superior performance in classification tasks and competitive results in anomaly detection.
KAUST has acquired a BM Pro plasma-enhanced chemical vapor deposition (PE-CVD) reactor from AIXTRON for wafer-scale deposition of graphene and carbon nanotubes. The reactor, capable of handling up to 4-inch substrates, will be used by Professor Pedro Da Costa's research team initially, before being opened up to other researchers at KAUST. AIXTRON's VP highlighted the system's uniformity, scalability, rapid heating, and plasma-based processing for growing graphene and nanotubes. Why it matters: This advanced tool enhances KAUST's research capabilities in carbon nanostructures, positioning the university as a leading center for materials science and nanotechnology research in the region.