Research On Lie Group Deep Structure Learning Algorithm

Lie group machine learning and deep structure learning have been taken much attention in recent years as new machine learning methods. In this paper, we integrate these two algorithms and put forward Lie group deep structure learning algorithm. The main research results are as follows:1) Introduction of Lie group machine learning and deep structure learning;2) How to process deep structure;3) A semi-supervised learning based on Lie group embedded in deep structure, the layered learning algorithm and heuristic learning algorithm are proposed, and their applications in the standard ORL database has been proved by the accuracy of face recognition.4) The experiments have show that Lie group deep structure learning algorithm can effectively identify face.On one hand, our research has enriched the content of Lie group machine learning. On the other hand, it can be seen from our research that our proposed algorithms perform well in human face identification and that deep structure can transform the complex problems into simple ones, which provides a novel way to the complex problems.