RL.TDl

Synopsis

• data TDl_Opts = TDl_Opts {
  • o_alpha :: TD_Number
  • o_gamma :: TD_Number
  • o_eps :: TD_Number
  • o_lambda :: TD_Number
  }
• type TD_Number = Double
• type Q s a = M s a TD_Number
• type Z s a = M s a TD_Number
• type V s a = HashMap s (a, TD_Number)
• emptyQ :: TD_Number -> Q s a
• toV :: (Bounded a, Enum a, Eq a, Hashable a, Eq s, Hashable s) => Q s a -> HashMap s TD_Number
• data TDl_State s a = TDl_State {
  • _tdl_q :: Q s a
  • _tdl_z :: Z s a
  }
• tdl_z :: forall s a. Lens' (TDl_State s a) (Z s a)
• tdl_q :: forall s a. Lens' (TDl_State s a) (Q s a)
• initialState :: Q s a -> TDl_State s a
• class (Eq s, Hashable s, Show s, Eq a, Hashable a, Enum a, Bounded a, Show a) => TDl_Problem pr m s a | pr -> m, pr -> s, pr -> a where
• queryQ :: (Hashable a, Hashable s, MonadState (TDl_State s a) f, Eq a, Eq s, Enum a, Bounded a) => s -> f [(a, TD_Number)]
• modifyQ :: (Hashable a, Hashable s, MonadState (TDl_State s a) m, Eq a, Eq s, Enum a, Bounded a) => t -> s -> a -> (TD_Number -> TD_Number) -> m ()
• listZ :: (TDl_Problem pr m s a, MonadTrans t, MonadState (TDl_State s a) (t m), Monad m) => pr -> s -> a -> ((s, a, TD_Number) -> t m b) -> t m ()
• modifyZ :: (Hashable a, Hashable s, MonadState (TDl_State s a) m, Eq a, Eq s, Enum a, Bounded a) => t -> s -> a -> (TD_Number -> TD_Number) -> m ()
• action :: (TDl_Problem pr m1 s1 a, MonadRnd g m, Hashable s, MonadState (TDl_State s a) m, Eq s) => pr -> s -> TD_Number -> m (a, TD_Number)
• transition :: (TDl_Problem pr m b a, MonadTrans t, MonadState (TDl_State b a) (t m), Monad m) => pr -> b -> a -> t m b
• getQ :: (Hashable s, Hashable a, MonadState (TDl_State s a) f, Eq s, Eq a, Enum a, Bounded a) => s -> a -> f TD_Number
• tdl_learn :: (MonadRnd g m, TDl_Problem pr m s a) => TDl_Opts -> Q s a -> s -> pr -> m (s, Q s a)
• qlw_learn :: (MonadRnd g m, TDl_Problem pr m s a) => TDl_Opts -> Q s a -> s -> pr -> m (s, Q s a)

Documentation