Does my machine learning approach make sense

Hey, So I want to do the following: I have multivariate time series data from sensors. I want to feed the raw, unlabeled data into a deep learning model (Thinking of deep belief nets right now). I hope that in the output layer of the rbm, there will be features of the time series. With every additional rbm, i will learn features of higher level . Those features will be used to represent the time series. Now some questions: 1) In such a case, how do you construct the input ? im thinking that every visible unit is a part of one sensor signal (with window size w). 2) Is a feature in the output an actual part of the time series? Or am I misunderstanding this. 3) Does anyone…

Link to Full Article: Does my machine learning approach make sense

Pin It on Pinterest

Share This