A multi-armed bandit (MAB) environment for the gymnasium API. One-armed Bandit is a reference to slot machines, and Buffalo is a reference to one such slot machine that I am fond of. MABs are an excellent playground for theoretical exercise and debugging of RL agents as they provide an environment that can be reasoned about easily. It helped me once to step back and write an MAB to debug my DQN agent. But there was a lack of native gymnasium environments, so I wrote Buffalo, an easy-to-use environment that it might help someone else.
Default multi-armed bandit environment. Arm center values are drawn from a normal distribution (0, arms). When an arm is pulled, a random value is drawn from a normal distribution (0, 1) and added to the chosen arm center value. This is not intended to be challenging for an agent but easy for the debugger to reason about.
This serves as a contextual bandit implementation. It is a k-armed bandit with n states. These states are indicated to the agent in the observation and the two states have different reward offsets for each arm. The goal of the agent is to learn and contextualize best action for a given state. This is a good stepping stone to Markov Decision Processes.
This module had an extra parameter, pace. By default (None), a new state is chosen for every step of the environment. It can be set to any integer to determine how many steps between randomly choosing a new state. Of course, transitioning to a new state is not guaranteed as the next state is random.
This serves as a stateful bandit implementation. There is a pervasive rumor that slot machine manufacturers put in a secret sequence of bets which trigger a large reward or the jackpot. It is almost certainly not true in the real world but it is here. A sequence of actions gives the max reward. The sequence is randomly chosen on environment setup and indicated in the info of reset. Not all sequences are aliased and this may be an important thing to check in an implementation. Therefore, there is a rudimentary algorithm to force aliasing included.
Install via pip and import buffalo_gym along with gymnasium.
import gymnasium
import buffalo_gym
env = gym.make("Buffalo-v0")