ABSTRACT
Automated blinds are installed on windows in a room to control the amount of heat and light transmitted into the room. The blinds are raised or lowered using motors according to the amount of daylight available. Autoregression is frequently used in predictions with time series. Sensors and other sources produce data at regular intervals. As the number of variables increases, it becomes easier to fit all the training data by adjusting the weight factors. In ridge regression, the norm of the weight vector is minimized along with the error function. In many applications, output variables depend on the input variables through non-linear relationships. Regression is used where the exact analytical model for predicting an output variable is not available or reliable. A loss function, typically using mean square error, is used to update the weights in the network such that the predicted output matches the output in the training data.
