Deep Learning in Time Series Analysis

2 Learning Theory

Learning in its broad sense implies a process, through which similarities and dissimilarities of various classes of previously prepared data, sometimes called training data, are found, and an appropriate decision is made regarding it’s implementation. In the context of computer sciences, learning is performed through a systematic procedure, known as the learning process. Here one should pay attention to the methodological differences in learning process between the conventional methods and the more recent forms of deep learning methods, especially when it comes to the time series analysis. In most of the conventional methods, a level of processing is involved, in which a mathematical or sometimes a statistical mapping is applied ...