Development and Application of an ICA algorithm for quite high-dimensional signals

虽然人脑可以从大量信号的复杂数据中快速学习到重要特征，如自然场景，并识别它们，但计算机要花很长的计算时间来做类似的处理。在这个项目中，以前的方法已被改进的统计和信息理论分析和计算时间已大大减少。最后，通过对自然场景等实际数据的数值实验，验证了该方法的有效性。

Though human brains can learn quickly the significant features from complicated data of a quite large number of signals such as natural scenes and recognize them, computers spend quite long computation time on doing the similar processing. In this project, the previous method has been improved by the statistical and information-theoretical analyses and the computation time has been reduced drastically. Besides, the validity of the proposed method is verified by numerical experiments on practical data such as natural scenes.