Skip to content
ml-recall

#reward-shaping

2 questions from RL Fundamentals

Timed test

2 questions · 2 min · score and explanations at the end

Take a test
Q 1 / 2Score 0 / 0

Question 1: Single choice

mediumAI-checkedRL Fundamentals

You train an agent to play Minesweeper. Each move ends in one of four ways: the agent wins (it cleared the last safe square), it makes an invalid move (uncovering a square already uncovered, which ends the game as a loss), it detonates a mine (a loss), or it clears a safe square and the game goes on. Which rewards (win / invalid / mine / safe square) make maximizing the cumulative reward the same as trying to win, whatever the size of the board?

Choose one option