Function windy_grid_penalized

rl/grid_world.py:265–333

(step_cost=-0.1)

def windy_grid_penalized(step_cost=-0.1):
  g = WindyGrid(3, 4, (2, 0))
  rewards = {
    (0, 0): step_cost,
    (0, 1): step_cost,
    (0, 2): step_cost,
    (1, 0): step_cost,
    (1, 2): step_cost,
    (2, 0): step_cost,
    (2, 1): step_cost,
    (2, 2): step_cost,
    (2, 3): step_cost,
    (0, 3): 1,
    (1, 3): -1
  }
  actions = {
    (0, 0): ('D', 'R'),
    (0, 1): ('L', 'R'),
    (0, 2): ('L', 'D', 'R'),
    (1, 0): ('U', 'D'),
    (1, 2): ('U', 'D', 'R'),
    (2, 0): ('U', 'R'),
    (2, 1): ('L', 'R'),
    (2, 2): ('L', 'R', 'U'),
    (2, 3): ('L', 'U'),
  }

  # p(s' | s, a) represented as:
  # KEY: (s, a) --> VALUE: {s': p(s' | s, a)}
  probs = {
    ((2, 0), 'U'): {(1, 0): 1.0},
    ((2, 0), 'D'): {(2, 0): 1.0},
    ((2, 0), 'L'): {(2, 0): 1.0},
    ((2, 0), 'R'): {(2, 1): 1.0},
    ((1, 0), 'U'): {(0, 0): 1.0},
    ((1, 0), 'D'): {(2, 0): 1.0},
    ((1, 0), 'L'): {(1, 0): 1.0},
    ((1, 0), 'R'): {(1, 0): 1.0},
    ((0, 0), 'U'): {(0, 0): 1.0},
    ((0, 0), 'D'): {(1, 0): 1.0},
    ((0, 0), 'L'): {(0, 0): 1.0},
    ((0, 0), 'R'): {(0, 1): 1.0},
    ((0, 1), 'U'): {(0, 1): 1.0},
    ((0, 1), 'D'): {(0, 1): 1.0},
    ((0, 1), 'L'): {(0, 0): 1.0},
    ((0, 1), 'R'): {(0, 2): 1.0},
    ((0, 2), 'U'): {(0, 2): 1.0},
    ((0, 2), 'D'): {(1, 2): 1.0},
    ((0, 2), 'L'): {(0, 1): 1.0},
    ((0, 2), 'R'): {(0, 3): 1.0},
    ((2, 1), 'U'): {(2, 1): 1.0},
    ((2, 1), 'D'): {(2, 1): 1.0},
    ((2, 1), 'L'): {(2, 0): 1.0},
    ((2, 1), 'R'): {(2, 2): 1.0},
    ((2, 2), 'U'): {(1, 2): 1.0},
    ((2, 2), 'D'): {(2, 2): 1.0},
    ((2, 2), 'L'): {(2, 1): 1.0},
    ((2, 2), 'R'): {(2, 3): 1.0},