Multi Armed Bandits

Personal project on multi-armed bandits.

Content

• Bandits: GaussianBandit
• Policies: Fixed, Greedy, Random, EpsilonGreedy, UCB, ThompsonSampling

Source: Reinforcement learning, Chapter 2, by Richard S. Sutton and Andrew G.Barto.

Install

This project was developed with python 3.7.7.

• Create virtual environment python -m venv venv
• Source:
  • Linux/macos source venv/bin/activate
  • Windows venv\Scripts\activate
• Upgrade pip python -m pip install --upgrade pip
• python -m pip install -U pip setuptools wheel
• Install requirements python -m pip install -r requirements.txt
• Install tool python -m pip install -e .

Example notebook explanation

The repo has an example notebook with some plots as discussed in the book Reinforcement Learning Chapter 2.