Skip to content

EpsilonGreedy

\(\varepsilon\)-greedy bandit policy.

Performs arm selection by using an \(\varepsilon\)-greedy bandit strategy. An arm is selected at each step. The best arm is selected (1 - \(\varepsilon\))% of the time.

Selection bias is a common problem when using bandits. This bias can be mitigated by using burn-in phase. Each model is given the chance to learn during the first burn_in steps.

Parameters

  • epsilon

    Type → float

    The probability of exploring.

  • decay

    Default → 0.0

    The decay rate of epsilon.

  • reward_obj

    Default → None

    The reward object used to measure the performance of each arm. This can be a metric, a statistic, or a distribution.

  • burn_in

    Default → 0

    The number of steps to use for the burn-in phase. Each arm is given the chance to be pulled during the burn-in phase. This is useful to mitigate selection bias.

  • seed

    Type → int | None

    Default → None

    Random number generator seed for reproducibility.

Attributes

  • current_epsilon

    The value of epsilon after factoring in the decay rate.

  • ranking

    Return the list of arms in descending order of performance.

Examples

import gymnasium as gym
from river import bandit
from river import stats

env = gym.make(
    'river_bandits/CandyCaneContest-v0'
)
_ = env.reset(seed=42)
_ = env.action_space.seed(123)

policy = bandit.EpsilonGreedy(epsilon=0.9, seed=101)

metric = stats.Sum()
while True:
    arm = policy.pull(range(env.action_space.n))
    observation, reward, terminated, truncated, info = env.step(arm)
    policy.update(arm, reward)
    metric.update(reward)
    if terminated or truncated:
        break

metric

Sum: 775.

Methods

pull

Pull arm(s).

This method is a generator that yields the arm(s) that should be pulled. During the burn-in phase, all the arms that have not been pulled enough times are yielded. Once the burn-in phase is over, the policy is allowed to choose the arm(s) that should be pulled. If you only want to pull one arm at a time during the burn-in phase, simply call next(policy.pull(arms)).

Parameters

  • arm_ids — 'list[ArmID]'

Returns

ArmID: A single arm.

update

Update an arm's state.

Parameters

  • arm_id
  • reward_args
  • reward_kwargs

  1. ε-Greedy Algorithm - The Multi-Armed Bandit Problem and Its Solutions - Lilian Weng ↩