Premium Content

Multi-Armed Bandits

Exploration and exploitation without state

This chapter requires a subscription to access.

What you'll unlock:

  • 1. The k-Armed Bandit Problem
  • 2. Action-Value Methods and ε-Greedy
  • 3. Upper-Confidence-Bound (UCB)
  • 4. Gradient Bandit Algorithms
  • 5. Thompson Sampling
  • 6. Contextual Bandits and LinUCB
Subscribe to Unlock

Already have an account? Sign in