Configure a practice, examination, calculation, scenario, or rapid-fire session. Progress is saved locally in your browser.
Quiz setup
Choose topics, difficulty, scoring, and feedback behavior.
70%
Untimed
Question palette
Session
Answered0
Bookmarked0
XP preview0
Formula reference
Return: Gt = Rt+1 + γRt+2 + γ²Rt+3 + …
Expected reward: E[R] = Σpiri
Success rate: successes ÷ episodes × 100
Greedy value: V(s) = maxaQ(s,a)
Performance report
Answer review
Agent-environment simulator
Move through a grid, collect rewards, avoid penalties, and inspect the trajectory.
Keyboard: arrow keys move the simulator when the quiz is not active.
0.90
Steps0
Reward0
Return0