Problem-07 You are asked to generate an Auto Regressive (AR) model and then create an RNN (such as LSTM) that predicts it. Generate samples of an Auto Regressive model of the form: X, = a;X 1 + a2X,2 + a3X,3 + U, where a;= 0.6, a;= -0.5, a;= -0.2 and U, is independent-identically distributed Uniform in 0, 0.1). Now train an RNN that predicts the sequence. Apply the training algorithm on new samples and calculate the averaged cost square error cost function. (a) Investigate different RNN structures and different number of training samples. (Important: You are not allowed to use your model knowledge when you design the RNN.) Problem-07 You are asked to generate an Auto Regressive (AR) model and then create an RNN (such as LSTM) that predicts it. Generate samples of an Auto Regressive model of the form: X, = a;X 1 + a2X,2 + a3X,3 + U, where a;= 0.6, a;= -0.5, a;= -0.2 and U, is independent-identically distributed Uniform in 0, 0.1). Now train an RNN that predicts the sequence. Apply the training algorithm on new samples and calculate the averaged cost square error cost function. (a) Investigate different RNN structures and different number of training samples. (Important: You are not allowed to use your model knowledge when you design the RNN.)