Indicate true or false.

a$)(5$ points$)$ Linear regression minimizes the total error ${\displaystyle \underset{i}{\overset{?}{}}}(y_i-{}^T{x}_i).[]$

b$)(5$ points$)$ Linear regression does not work for three$-$class classification problems. $[]$

c$)(5$ points$)$ The state of a hidden Markov model can be directly observed. $[]$

d$)(5$ points$)$ A deep neural network is a neural network with multiple hidden units. $[]$

e$)(5$ points$)$ The multi$-$arm bandit problem has an environment with a singleton state space. $[]$

f$)(5$ points$)$ Monte$-$Carlo policy evaluation does not require knowledge of transition

probabilities. $[]$