BasicBernoulliBandit {contextual} | R Documentation |
Context-free Bernoulli or Binary multi-armed bandit.
Simulates k
Bernoulli arms where each arm issues a reward of one with
uniform probability p
, and otherwise a reward of zero.
In a bandit scenario, this can be used to simulate a hit or miss event, such as if a user clicks on a headline, ad, or recommended product.
bandit <- BasicBernoulliBandit$new(weights)
weights
numeric vector; probability of reward values for each of the bandit's k
arms
new(weights)
generates and instantializes a new BasicBernoulliBandit
instance.
get_context(t)
argument:
t
: integer, time step t
.
returns a named list
containing the current d x k
dimensional matrix context$X
,
the number of arms context$k
and the number of features context$d
.
get_reward(t, context, action)
arguments:
t
: integer, time step t
.
context
: list, containing the current context$X
(d x k context matrix),
context$k
(number of arms) and context$d
(number of context features)
(as set by bandit
).
action
: list, containing action$choice
(as set by policy
).
returns a named list
containing reward$reward
and, where computable,
reward$optimal
(used by "oracle" policies and to calculate regret).
Core contextual classes: Bandit
, Policy
, Simulator
,
Agent
, History
, Plot
Bandit subclass examples: BasicBernoulliBandit
, ContextualLogitBandit
,
OfflineReplayEvaluatorBandit
Policy subclass examples: EpsilonGreedyPolicy
, ContextualLinTSPolicy
## Not run: horizon <- 100 sims <- 100 policy <- EpsilonGreedyPolicy$new(epsilon = 0.1) bandit <- BasicBernoulliBandit$new(weights = c(0.6, 0.1, 0.1)) agent <- Agent$new(policy,bandit) history <- Simulator$new(agent, horizon, sims)$run() plot(history, type = "cumulative", regret = TRUE) ## End(Not run)