rl

type Interpreter interface {
	RewardFunc(State, *strategies.Action) float64
	Interpret(*types.Event, *strategies.Context) State
	Actions(State, *strategies.Context) []*strategies.Action
}

type Policy interface {
	NextAction(*StateActionMap, State, []*strategies.Action) (*strategies.Action, bool)
}

type RLStrategy struct {
	*types.BaseService
	// contains filtered or unexported fields
}

type RLStrategyConfig struct {
	Alpha       float64
	Gamma       float64
	Interpreter Interpreter
	Policy      Policy
}

type SimpleInterpreter struct {
	// contains filtered or unexported fields
}

type SimpleInterpreterState struct {
	// contains filtered or unexported fields
}

type SoftMaxPolicy struct {
}

type State interface {
	Hash() string
}

type StateActionMap struct {
	// contains filtered or unexported fields
}