← All explorers
Loading explorer…
Read the theory: rl-basics
Read the theory: rl-basics
Four arms, one is best. Compare greedy, ε-greedy, and UCB on the same reward draws — and watch which strategies keep regret low.
Each line averages cumulative pseudo-regret over 200 runs: the sum of gaps between the best arm's mean reward and each chosen arm's mean. Lower is better. Greedy can get stuck, constant ε-greedy keeps exploring at a long-run cost, and UCB directs exploration using uncertainty. Inspect both single runs and averages.
Illustrative model for building intuition, not a live system.