Usage and Examples

Notebooks

These notebooks demonstrate various features and use cases of the library.

Notebooks:

• Thompson Sampling and the Bernoulli Bandit
  • Optimizing Click-Through Rates
  • Simulating the Problem
  • Setting up the learning agent
• Contextual Bandits for Count Data
  • Maximizing Web Storefront Transactions per Week
  • Simulating the Problem
  • Setting up the learner
• A Bayesian Take on Linear Bandits and Disjoint LinUCB
  • What should we recommend on the “Featured” section of our website?
  • Simulating the problem
• Delayed and non-stationary multi-armed bandit problems
  • How can we implement a geographic pricing strategy?
  • Simulating the problem
• Bandit Persistence Recipes