Q32:In Q-learning, suppose we have a Q-table (whose values a…
Q32:In Q-learning, suppose we have a Q-table (whose values are Q(s, a)) as follows. The action space is {a₁, a₂, a₃} and state space is {s₁, s₂, s₃, s₄}. (7 points) Which action should the agent take when sₜ = s₂ at time t? Why? s₁ s₂ s₃ s₄ a₁ 1.20 2.40 2.50 1.47 a₂ 0.33 0.70 1.31 1.50 a₃ -0.90 -2.80 1.00 0.80