Value Members

1.  final def !=(arg0: Any): Boolean

   Definition Classes

   AnyRef → Any

2.  final def ##(): Int

   Definition Classes

   AnyRef → Any

3.  final def ==(arg0: Any): Boolean

   Definition Classes

   AnyRef → Any

4.  final def asInstanceOf[T0]: T0

   Definition Classes

   Any

5.  def byPolicy[Obs, A, R, M[_]](basePolicy: Policy[Obs, A, R, Cat, M], targetPolicy: Policy[Obs, A, R, Cat, M]): (State[Obs, A, R, M], A, R) ⇒ Weight
6.  def clone(): AnyRef

   Attributes

   protected[lang]

   Definition Classes

   AnyRef

   Annotations

   @throws( ... ) @native()

7.  def constant[Obs, A, R, M[_]]: (SARS[Obs, A, R, M]) ⇒ Weight
8.  final def eq(arg0: AnyRef): Boolean

   Definition Classes

   AnyRef

9.  def equals(arg0: Any): Boolean

   Definition Classes

   AnyRef → Any

10.  def finalize(): Unit

    Attributes

    protected[lang]

    Definition Classes

    AnyRef

    Annotations

    @throws( classOf[java.lang.Throwable] )

11.  final def getClass(): Class[_]

    Definition Classes

    AnyRef → Any

    Annotations

    @native()

12.  def hashCode(): Int

    Definition Classes

    AnyRef → Any

    Annotations

    @native()

13.  final def isInstanceOf[T0]: Boolean

    Definition Classes

    Any

14.  final def ne(arg0: AnyRef): Boolean

    Definition Classes

    AnyRef

15.  final def notify(): Unit

    Definition Classes

    AnyRef

    Annotations

    @native()

16.  final def notifyAll(): Unit

    Definition Classes

    AnyRef

    Annotations

    @native()

17.  def processTrajectory[Obs, A, R, G, M[_]](trajectory: Trajectory[Obs, A, R, M], valueFn: ActionValueFn[Obs, A, G], agg: MonoidAggregator[SARS[Obs, A, R, M], G, Option[G]]): ActionValueFn[Obs, A, G]

    So if you have G, your return...

    So if you have G, your return... okay, this is a version that tracks the weights, but doesn't give you a nice way to push the weights back. What if we make the weight part of G? Try that in the next fn.

    This is a full monte carlo trajectory tracker that's able to do off-policy control. The behavior policy does NOT change at all, but that's okay, I guess. We're going to have to solve that now. Presumably if you're updating a value function at any point you could get a new agent.

18.  def processTrajectorySimple[Obs, A, R, G, M[_]](trajectory: Trajectory[Obs, A, R, M], valueFn: ActionValueFn[Obs, A, G], agg: MonoidAggregator[R, G, G]): ActionValueFn[Obs, A, G]

    This is a simpler version that doesn't do any weighting.

    This is a simpler version that doesn't do any weighting. This should be equivalent to the more difficult one above, with a constant weight of 1 for everything.

19.  def sarsa[Obs, A, R, M[_], T](moment: Moment[Obs, A, R, M], tracker: Tracker[Obs, A, R, T, M])(implicit arg0: Monad[M]): M[(Moment[Obs, A, R, M], Trajectory[Obs, A, R, M])]
20.  final def synchronized[T0](arg0: ⇒ T0): T0

    Definition Classes

    AnyRef

21.  def toString(): String

    Definition Classes

    AnyRef → Any

22.  final def wait(): Unit

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )

23.  final def wait(arg0: Long, arg1: Int): Unit

    Definition Classes

    AnyRef

    Annotations

    @throws( ... )

24.  final def wait(arg0: Long): Unit

    Definition Classes

    AnyRef

    Annotations

    @throws( ... ) @native()

25.  def weighted[Obs, A, R, G, M[_]](agg: MonoidAggregator[R, G, G], fn: (SARS[Obs, A, R, M]) ⇒ Weight): MonoidAggregator[SARS[Obs, A, R, M], (G, Weight), Option[(G, Weight)]]
26.  object ShouldUpdateState extends Serializable
27.  object Tracker