{ "meta": { "policies": [ "zero_op", "greedy", "optimal_replay" ], "tiers": [ 1, 2 ], "n_seeds": 5, "duration_s": 4.63 }, "summary": { "tier1/zero_op": { "gap": { "n": 5, "mean": 4.478881827096123, "median": 4.202050754119967, "stdev": 0.9217149361737852, "min": 3.3781262294160617, "max": 5.444144463465587 }, "terminal": { "n": 5, "mean": 0.18673441916665726, "median": 0.19223188070743255, "stdev": 0.031493829949386834, "min": 0.15517963721474534, "max": 0.2284082156611041 }, "agent_cost": { "n": 5, "mean": 4196.7, "median": 4284.5, "stdev": 219.84841368543007, "min": 3805.0, "max": 4325.5 }, "optimal_cost": { "n": 5, "mean": 782.235, "median": 769.3520000000002, "stdev": 130.6360389077227, "min": 663.0049999999998, "max": 978.6150000000001 } }, "tier1/greedy": { "gap": { "n": 5, "mean": 1.5936242753244134, "median": 1.5268617280357517, "stdev": 0.2481291406117591, "min": 1.3244146063569426, "max": 1.9379605924295358 }, "terminal": { "n": 5, "mean": 0.3883272601529545, "median": 0.39574781196173614, "stdev": 0.036219638080110333, "min": 0.3403721624370236, "max": 0.43021584757949055 }, "agent_cost": { "n": 5, "mean": 2006.4966, "median": 1964.626, "stdev": 183.88109805333443, "min": 1822.415, "max": 2274.707 }, "optimal_cost": { "n": 5, "mean": 782.235, "median": 769.3520000000002, "stdev": 130.6360389077227, "min": 663.0049999999998, "max": 978.6150000000001 } }, "tier1/optimal_replay": { "gap": { "n": 5, "mean": 0.06565162880132012, "median": 0.05405899465904312, "stdev": 0.038793269984578244, "min": 0.02477991855837064, "max": 0.11809865687287457 }, "terminal": { "n": 5, "mean": 0.9393779029200451, "median": 0.948713501869475, "stdev": 0.03382812622630545, "min": 0.8943754594937315, "max": 0.9758192777691916 }, "agent_cost": { "n": 5, "mean": 830.075, "median": 799.052, "stdev": 112.6100206975383, "min": 730.704, "max": 1002.865 }, "optimal_cost": { "n": 5, "mean": 782.235, "median": 769.3520000000002, "stdev": 130.6360389077227, "min": 663.0049999999998, "max": 978.6150000000001 } }, "tier2/zero_op": { "gap": { "n": 5, "mean": 1.013266828092146, "median": 0.8737750480918441, "stdev": 0.323903443066573, "min": 0.7683094607125767, "max": 1.5397854975061123 }, "terminal": { "n": 5, "mean": 0.5059878294124428, "median": 0.5336819918796277, "stdev": 0.07244476047091011, "min": 0.39373403816264346, "max": 0.5655118757307496 }, "agent_cost": { "n": 5, "mean": 62846.7, "median": 63150.0, "stdev": 4491.170805035142, "min": 56940.5, "max": 67281.0 }, "optimal_cost": { "n": 5, "mean": 31967.512, "median": 33857.195999999996, "stdev": 6062.466152161799, "min": 22419.413, "max": 37806.142 } }, "tier2/greedy": { "gap": { "n": 5, "mean": 0.34603178178836785, "median": 0.2732357392047041, "stdev": 0.1731178523338452, "min": 0.18901783833486985, "max": 0.6096376385947304 }, "terminal": { "n": 5, "mean": 0.7521874151378729, "median": 0.7854005108469748, "stdev": 0.09029988955815418, "min": 0.6212578384244513, "max": 0.8410302753745267 }, "agent_cost": { "n": 5, "mean": 42242.8058, "median": 42867.461, "stdev": 4211.305052673067, "min": 36087.131, "max": 46481.363 }, "optimal_cost": { "n": 5, "mean": 31967.512, "median": 33857.195999999996, "stdev": 6062.466152161799, "min": 22419.413, "max": 37806.142 } }, "tier2/optimal_replay": { "gap": { "n": 5, "mean": 0.018469179466918097, "median": 0.016450632829818797, "stdev": 0.006732061234245376, "min": 0.012397659865276715, "max": 0.029369636038196038 }, "terminal": { "n": 5, "mean": 0.9818998850063403, "median": 0.9838156106174877, "stdev": 0.006456151657885251, "min": 0.9714683287616361, "max": 0.9877541598951083 }, "agent_cost": { "n": 5, "mean": 32527.572000000004, "median": 34276.946, "stdev": 6007.788721970215, "min": 23077.863, "max": 38339.192 }, "optimal_cost": { "n": 5, "mean": 31967.512, "median": 33857.195999999996, "stdev": 6062.466152161799, "min": 22419.413, "max": 37806.142 } } }, "rollouts": [ { "policy": "zero_op", "tier": 1, "seed": 0, "agent_cost": 4284.5, "optimal_cost": 978.6150000000001, "gap": 3.3781262294160617, "terminal": 0.2284082156611041, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 1, "seed": 1, "agent_cost": 3805.0, "optimal_cost": 769.3520000000002, "gap": 3.945720554440619, "terminal": 0.2021950065703023, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 1, "seed": 2, "agent_cost": 4272.5, "optimal_cost": 663.0049999999998, "gap": 5.444144463465587, "terminal": 0.15517963721474534, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 1, "seed": 3, "agent_cost": 4325.5, "optimal_cost": 831.4990000000003, "gap": 4.202050754119967, "terminal": 0.19223188070743255, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 1, "seed": 4, "agent_cost": 4296.0, "optimal_cost": 668.7039999999998, "gap": 5.4243671340383806, "terminal": 0.15565735567970193, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "greedy", "tier": 1, "seed": 0, "agent_cost": 2274.707, "optimal_cost": 978.6150000000001, "gap": 1.3244146063569426, "terminal": 0.43021584757949055, "cumulative_dense": 0.4, "steps": 58 }, { "policy": "greedy", "tier": 1, "seed": 1, "agent_cost": 1869.652, "optimal_cost": 769.3520000000002, "gap": 1.4301646060580846, "terminal": 0.41149475945256136, "cumulative_dense": 0.4, "steps": 50 }, { "policy": "greedy", "tier": 1, "seed": 2, "agent_cost": 1822.415, "optimal_cost": 663.0049999999998, "gap": 1.7487198437417526, "terminal": 0.3638057193339606, "cumulative_dense": 0.4, "steps": 44 }, { "policy": "greedy", "tier": 1, "seed": 3, "agent_cost": 2101.083, "optimal_cost": 831.4990000000003, "gap": 1.5268617280357517, "terminal": 0.39574781196173614, "cumulative_dense": 0.4, "steps": 52 }, { "policy": "greedy", "tier": 1, "seed": 4, "agent_cost": 1964.626, "optimal_cost": 668.7039999999998, "gap": 1.9379605924295358, "terminal": 0.3403721624370236, "cumulative_dense": 0.4, "steps": 55 }, { "policy": "optimal_replay", "tier": 1, "seed": 0, "agent_cost": 1002.865, "optimal_cost": 978.6150000000001, "gap": 0.02477991855837064, "terminal": 0.9758192777691916, "cumulative_dense": 0.4, "steps": 53 }, { "policy": "optimal_replay", "tier": 1, "seed": 1, "agent_cost": 799.052, "optimal_cost": 769.3520000000002, "gap": 0.038603916022834556, "terminal": 0.9628309546812978, "cumulative_dense": 0.4, "steps": 53 }, { "policy": "optimal_replay", "tier": 1, "seed": 2, "agent_cost": 741.305, "optimal_cost": 663.0049999999998, "gap": 0.11809865687287457, "terminal": 0.8943754594937315, "cumulative_dense": 0.4, "steps": 58 }, { "policy": "optimal_replay", "tier": 1, "seed": 3, "agent_cost": 876.449, "optimal_cost": 831.4990000000003, "gap": 0.05405899465904312, "terminal": 0.948713501869475, "cumulative_dense": 0.4, "steps": 56 }, { "policy": "optimal_replay", "tier": 1, "seed": 4, "agent_cost": 730.704, "optimal_cost": 668.7039999999998, "gap": 0.09271665789347773, "terminal": 0.9151503207865291, "cumulative_dense": 0.4, "steps": 57 }, { "policy": "zero_op", "tier": 2, "seed": 0, "agent_cost": 63150.0, "optimal_cost": 30002.385, "gap": 1.104832665803069, "terminal": 0.47509714964370614, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 2, "seed": 1, "agent_cost": 66992.0, "optimal_cost": 35752.424, "gap": 0.8737750480918441, "terminal": 0.5336819918796277, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 2, "seed": 2, "agent_cost": 59870.0, "optimal_cost": 33857.195999999996, "gap": 0.7683094607125767, "terminal": 0.5655118757307496, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 2, "seed": 3, "agent_cost": 56940.5, "optimal_cost": 22419.413, "gap": 1.5397854975061123, "terminal": 0.39373403816264346, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "zero_op", "tier": 2, "seed": 4, "agent_cost": 67281.0, "optimal_cost": 37806.142, "gap": 0.7796314683471273, "terminal": 0.5619140916454871, "cumulative_dense": 0.4, "steps": 30 }, { "policy": "greedy", "tier": 2, "seed": 0, "agent_cost": 42867.461, "optimal_cost": 30002.385, "gap": 0.4288017769254013, "terminal": 0.6998871475033243, "cumulative_dense": 0.4, "steps": 41 }, { "policy": "greedy", "tier": 2, "seed": 1, "agent_cost": 45521.264, "optimal_cost": 35752.424, "gap": 0.2732357392047041, "terminal": 0.7854005108469748, "cumulative_dense": 0.4, "steps": 40 }, { "policy": "greedy", "tier": 2, "seed": 2, "agent_cost": 40256.81, "optimal_cost": 33857.195999999996, "gap": 0.18901783833486985, "terminal": 0.8410302753745267, "cumulative_dense": 0.4, "steps": 39 }, { "policy": "greedy", "tier": 2, "seed": 3, "agent_cost": 36087.131, "optimal_cost": 22419.413, "gap": 0.6096376385947304, "terminal": 0.6212578384244513, "cumulative_dense": 0.4, "steps": 42 }, { "policy": "greedy", "tier": 2, "seed": 4, "agent_cost": 46481.363, "optimal_cost": 37806.142, "gap": 0.22946591588213358, "terminal": 0.8133613035400877, "cumulative_dense": 0.4, "steps": 39 }, { "policy": "optimal_replay", "tier": 2, "seed": 0, "agent_cost": 30603.285, "optimal_cost": 30002.385, "gap": 0.020028407741584594, "terminal": 0.9803648529888211, "cumulative_dense": 0.4000000000000001, "steps": 69 }, { "policy": "optimal_replay", "tier": 2, "seed": 1, "agent_cost": 36340.574, "optimal_cost": 35752.424, "gap": 0.016450632829818797, "terminal": 0.9838156106174877, "cumulative_dense": 0.4000000000000001, "steps": 61 }, { "policy": "optimal_replay", "tier": 2, "seed": 2, "agent_cost": 34276.946, "optimal_cost": 33857.195999999996, "gap": 0.012397659865276715, "terminal": 0.9877541598951083, "cumulative_dense": 0.4000000000000001, "steps": 60 }, { "policy": "optimal_replay", "tier": 2, "seed": 3, "agent_cost": 23077.863, "optimal_cost": 22419.413, "gap": 0.029369636038196038, "terminal": 0.9714683287616361, "cumulative_dense": 0.4000000000000001, "steps": 69 }, { "policy": "optimal_replay", "tier": 2, "seed": 4, "agent_cost": 38339.192, "optimal_cost": 37806.142, "gap": 0.014099560859714352, "terminal": 0.9860964727686488, "cumulative_dense": 0.4000000000000001, "steps": 63 } ] }