LSTM Gradient [4pts] Here, you'll derive the Backprop Through Time equations for the univariate version of the Long-Term Short-Term Memory (LSTM) architecture For reference, here are the computations it performs (wo)) g(t) = tanh(wgzX(t) + tVghh(t-1)) (t-1) dt) = f(t) c(t-1) + (t)g(t) tan(l) (a) [3pts] Derive the Backprop Through Time equations for the activations and the gates h(t) = f(t) = You don't need to vectorize anything or factor out any repeated subexpressions (b) lpt] Derive the BPTT equation for the weight wiz (The other weight matrices are basically the same, so we won't make you write those out LSTM Gradient [4pts] Here, you'll derive the Backprop Through Time equations for the univariate version of the Long-Term Short-Term Memory (LSTM) architecture For reference, here are the computations it performs (wo)) g(t) = tanh(wgzX(t) + tVghh(t-1)) (t-1) dt) = f(t) c(t-1) + (t)g(t) tan(l) (a) [3pts] Derive the Backprop Through Time equations for the activations and the gates h(t) = f(t) = You don't need to vectorize anything or factor out any repeated subexpressions (b) lpt] Derive the BPTT equation for the weight wiz (The other weight matrices are basically the same, so we won't make you write those out