Academy of Mathematics and Systems Science, CAS
Colloquia & Seminars

Speaker:	Xiaotong Shen 教授，University of Minnesota
Inviter:
Title:	Embedding Learning
Time & Venue:	2019.5.27 10:00-11:00 N602
Abstract:	Numerical embedding has become one standard technique for processing and analyzing unstructured data that cannot be expressed in a predefined fashion, for example, text and video-caption data. It stores the main characteristics of data by mapping it onto a numerical vector. An embedding is often unsupervised and constructed by transfer learning from large-scale unannotated data. Given an embedding, a downstream learning method, referred to as a two-stage method, is applicable to unstructured data. In this article, we introduce a novel framework of embedding learning to deliver a higher learning accuracy than the two-stage method while identifying an optimal learning-adaptive embedding. Particularly, we propose a concept of minimal sufficient learning-adaptive embeddings, based on which we seek an optimal one to maximize the learning accuracy subject to an embedding loss constraint. Some computational and theoretical aspects of embedding learning will be discussed, in addition to examples based on Word2Vec, Doc2Vec, and local linear embeddings. Finally, we demonstrate the utility of embedding learning on two benchmarks in representational learning and sentiment analysis