UCB — Upper Confidence Bound

Fact

A picking rule that balances trying options with earning rewards under uncertainty.

In Plain Words

UCB is like choosing lunch at school. You grab the safe pizza, but still risk the mystery taco once.

It picks the choice with the best hopeful score. It shows up in slot-machine-style tests and recommendation feeds.

Related Concepts

• Exploration-Exploitation Tradeoff
• Thompson Sampling
• Reinforcement Learning

Exploration-Exploitation Tradeoff
UCB is a classic way to balance trying new choices and using known winners.

Thompson Sampling
UCB and Thompson Sampling both handle the choice between testing and earning.

RL
UCB can be used in RL to choose the next action.