Tabular prediction with α=0.1, γ=1, and every state initialised to V=0.5. A single episode is observed:
Ar=0Br=0Cr=1terminal
Apply constant-α Monte Carlo, V(St)←V(St)+α[Gt−V(St)], and TD(0), V(St)←V(St)+α[Rt+1+γV(St+1)−V(St)]. What is V(A) after this one episode under each method?
A
Monte Carlo 0.50, TD(0) 0.55
B
Monte Carlo 0.55, TD(0) 0.50
C
Monte Carlo 0.55, TD(0) 0.55
D
Monte Carlo 1.00, TD(0) 0.50
Sign in to answer questions and track your progress