Code QuizIntermediate

MDP Step Function

Spot the bug in an MDP step returning reward and terminal flag.

Codejavascript

// An MDP transition: given state+action, produce next state & reward
function step(state, action) {
  const nextState = transition(state, action);
  const reward = rewardFn(state, action);
  const done = isTerminal(state);
  return { reward, done };
}

What is the bug in this MDP step function?