AbstractModels across different application domains are exploding in size. This disproportionate increase in model sizes will quickly obstruct the sustainability of deep learning. From my personal view, functional clustering, a model derived to classify a population of variables into distinct modules, can serve as an alternative tool to compress and optimize large multimodal models in any size....
摘要:This is an summary talk for Moser-Trudinger inequality. In this talk, I first introduce the classical Moser-Trudinger inequalities in bounded domain and unbounded domain. Then I will present the work on the existence and no-existence for extremals of Moser-Trudinger inequality. The Trace-type Moser-Trudinger inequality in bounded and unbounded domain will be also discussed in this talk. F...