AnytimeToKnownHorizonAgentWrapper

class statrl.settings.bandits.stochastic.knownhorizon.wrappers.wrapper_anytime_knownhorizon.AnytimeToKnownHorizonAgentWrapper(learner)

Bases: BanditAgent

Run an anytime agent inside the known-horizon setting.

An anytime agent simply ignores the horizon, so the wrapper drops it.

Parameters:

learner (statrl.settings.bandits.stochastic.anytime.agent.BanditAgent) – The anytime agent to adapt.

See also

KnownHorizonToAnytimeAgentWrapper

The converse adaptation.

Examples

>>> from statrl.settings.bandits.stochastic.anytime.agents.IMED import IMED
>>> AnytimeToKnownHorizonAgentWrapper(IMED(3)).name
'IMED-anytime'

Methods

__init__(learner)
reset(horizon)	Reset the wrapped agent, discarding the horizon.
select_arm()	Delegate the choice to the wrapped agent.
update(arm, reward)	Forward the observation to the wrapped agent.

Attributes

policy

Policy of the wrapped agent.

property policy

Policy of the wrapped agent.

Raises:

AttributeError – If the wrapped agent has no policy. Only oracles define one; the experiment harness reads it to log the optimal policy.

reset(horizon)

Reset the wrapped agent, discarding the horizon.

Parameters:

horizon (int) – Ignored: an anytime agent must not depend on it.

select_arm()

Delegate the choice to the wrapped agent.

Returns:

Index of the selected arm.

Return type:

int

update(arm, reward)

Forward the observation to the wrapped agent.

Parameters:

• arm (int) – Index of the arm that was pulled.

• reward (float) – Reward observed for that arm.