Enum relearn::defs::agent::AgentDef

pub enum AgentDef {
    Random,
    TabularQLearning(TabularQLearningAgentConfig),
    BetaThompsonSampling(BetaThompsonSamplingAgentConfig),
    UCB1(UCB1AgentConfig),
    ActorCritic(Box<ActorCriticConfig<PolicyDef, PolicyUpdaterDef, CriticDef, CriticUpdaterDef>>),
    ResettingMeta(Box<AgentDef>),
}

Agent definition

Variants

Random

An agent that selects actions randomly.

TabularQLearning(TabularQLearningAgentConfig)

Epsilon-greedy tabular Q learning.

Tuple Fields of TabularQLearning

0: TabularQLearningAgentConfig

BetaThompsonSampling(BetaThompsonSamplingAgentConfig)

Thompson sampling of for Bernoulli rewards using Beta priors.

Assumes no relationship between states.

Tuple Fields of BetaThompsonSampling

0: BetaThompsonSamplingAgentConfig

UCB1(UCB1AgentConfig)

UCB1 agent from Auer 2002

Tuple Fields of UCB1

0: UCB1AgentConfig

ActorCritic(Box<ActorCriticConfig<PolicyDef, PolicyUpdaterDef, CriticDef, CriticUpdaterDef>>)

Torch actor-critic agent

Tuple Fields of ActorCritic

0: Box<ActorCriticConfig<PolicyDef, PolicyUpdaterDef, CriticDef, CriticUpdaterDef>>

ResettingMeta(Box<AgentDef>)

Applies a non-meta agent to a meta environment by resetting between trials

Tuple Fields of ResettingMeta

0: Box<AgentDef>

Trait Implementations

impl Clone for AgentDef

fn clone(&self) -> AgentDef

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for AgentDef

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl PartialEq<AgentDef> for AgentDef

fn eq(&self, other: &AgentDef) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &AgentDef) -> bool

This method tests for !=.

impl StructuralPartialEq for AgentDef