HSVI for DecPOMDP (Serial vs Simultaneous)
|Comparaison des méthodes
mabc_100_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 100 | 0.001 | 0.01 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 1.28028 | 3 | 0.00828585 | 90.738 | 90.7463 | 244 | 244 | 58 | 118 | 1 | 8253 |
| SER | 1.43325 | 15 | 0.00137701 | 90.7389 | 90.7403 | 814 | 814 | 127 | 297 | 1 | 8269 |
| SIM (Maxplan/Sawtooth) | 1.3624 | 5 | 0.0074837 | 90.707 | 90.7145 | 167 | 163 | 62 | 118 | 1 | 13544 |
| SER (Maxplan/Sawtooth) | 1.50633 | 21 | -0.329728 | 89.4216 | 89.0918 | 317 | 328 | 136 | 297 | 1 | 13550 |
| SIM (MaxplanWCSP/SawtoothLP) | 6.07254 | 3 | 0.00938187 | 90.7348 | 90.7442 | 114 | 109 | 31 | 118 | 1 | 10308 |
RESULTS (do not store graph)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM (Maxplan/Sawtooth) | 1.57218 | 3 | 0.00933716 | 90.7348 | 90.7442 | 114 | 115 | 1 | 118 | 0 | 13380 |
| SER (Maxplan/Sawtooth) | 2.22411 | 31 | 0.00686318 | 89.7545 | 89.7613 | 446 | 620 | 1 | 297 | 0 | 12592 |
recycling_10_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 10 | 0.1 | 0.1 | 0.1 |
RESULTS (store graph)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 5.56031 | 203 | 0.00998992 | 31.4 | 31.41 | 283 | 283 | 123 | 5 | 4 | 3813 |
| SER | 2.34171 | 279 | 0.00362873 | 31.755 | 31.7586 | 647 | 647 | 553 | 17 | 4 | 3461 |
| SIM (Maxplan/Sawtooth) | 1.82763 | 19 | -0.0135327 | 31.8639 | 31.8504 | 21 | 35 | 102 | 5 | 4 | 13709 |
| SER (Maxplan/Sawtooth) | 481.489 | 34465 | 0.0396951 | 31.7455 | 31.7852 | 103 | 198 | 319 | 17 | 4 | 13889 |
| SIM (MaxplanWCSP/SawtoothLP) | 2.63119 | 10 | -0.180332 | 31.8639 | 31.6836 | 17 | 21 | 14 | 5 | 4 | 10309 |
RESULTS (do not store graph)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM (Maxplan/Sawtooth) | 3.681 | 18 | 3.55271e-15 | 31.8639 | 31.8639 | 21 | 58 | 1 | 5 | 0 | 12435 |
| SER (Maxplan/Sawtooth) | 4.37389 | 75 | 3.55271e-15 | 31.8639 | 31.8639 | 112 | 664 | 1 | 17 | 0 | 12464 |
recycling_20_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 20 | 0.001 | 0.01 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 1072.45 | 14358 | 0.614851 | 62.6324 | 63.2473 | 40209 | 40209 | 18695 | 5 | 4 | 15528 |
| SER | 482.781 | 34465 | 0.614563 | 62.514 | 63.1286 | 145944 | 145944 | 470 | 17 | 4 | 15528 |
aligment_10_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 10 | 0.001 | 0.01 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 111.246 | 2552 | 4.26326e-14 | 91 | 91 | 3798 | 3798 | 13379 | 257 | 2 | 6752 |
| SER | 191.987 | 14298 | 0.00520833 | 91 | 91.0052 | 21576 | 21576 | 43993 | 1272 | 2 | 7707 |
| SIM (Maxplan/Sawtooth) | 1802.03 | 940 | 50.682 | 91 | 141.682 | 798 | 3009 | 21354 | 20214 | 2 | 14902 |
| SER (Maxplan/Sawtooth) | 1801.52 | 1229 | 51.4648 | 91 | 142.465 | 1734 | 8317 | 30898 | 22579 | 2 | 15154 |
| SIM (MaxplanWCSP/SawtoothLP) | 1802.31 | 91 | 17.1729 | 91 | 108.173 | 384 | 488 | 600 | 1304 | 2 | 15285 |
Mars_5_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 5 | 0.1 | 0.1 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 1801.67 | 0 | 69.3856 | -55 | 14.3856 | 0 | 0 | 1 | 95 | 0 | 7047 |
| SER | 1803.23 | 13 | 2.17971 | 11.8366 | 14.0163 | 51 | 51 | 1 | 369 | 0 | 6544 |
| SIM (Maxplan/Sawtooth) | 1801.45 | 32 | 1.97402 | 12.3755 | 14.3495 | 96 | 280 | 1 | 760 | 0 | 13915 |
| SER (Maxplan/Sawtooth) | 3481.49 | 0 | 69.3856 | -55 | 14.3856 | 5 | 0 | 1 | 375 | 0 | 14163 |
| SIM (MaxplanWCSP/SawtoothLP) | 1802.2 | 28 | 1.06659 | 13.2665 | 14.3331 | 36 | 113 | 1 | 174 | 0 | 14850 |
GridSmall_10_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 10 | 0.1 | 0.1 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 78.5667 | 17 | -0.0892865 | 6.35295 | 6.26367 | 66 | 66 | 1 | 629 | 0 | 7186 |
| SER | 6.6975 | 16 | -0.0781523 | 6.25727 | 6.17912 | 107 | 107 | 1 | 3818 | 0 | 7176 |
| SIM (Maxplan/Sawtooth) | 1801.64 | 112 | 0.458392 | 5.80714 | 6.26553 | 96 | 666 | 1 | 1287 | 0 | 14047 |
| SER (Maxplan/Sawtooth) | 1802.39 | 225 | 0.427174 | 5.76089 | 6.18807 | 401 | 2649 | 1 | 4235 | 0 | 14115 |
| SIM (MaxplanWCSP/SawtoothLP) | 1802.84 | 27 | 0.679946 | 5.74564 | 6.42559 | 81 | 110 | 1 | 31800 | 0 | 13423 |
tiger_4_3
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 4 | 0.01 | 0.01 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 1801.68 | 5 | 2.99098 | 4.02227 | 7.01325 | 8 | 8 | 1 | 5 | 0 | 8304 |
| SER | 422.387 | 136 | 0.00454273 | 4.02227 | 4.02681 | 173 | 173 | 1 | 20 | 0 | 7967 |
| SIM (Maxplan/Sawtooth) | 1998.33 | 4 | 3.49731 | 3.51594 | 7.01325 | 10 | 13 | 1 | 101 | 0 | 14666 |
| SER (Maxplan/Sawtooth) | 1801.68 | 1090 | 0.0891938 | 3.51594 | 3.60513 | 188 | 5516 | 1 | 194 | 0 | 14985 |
| SIM (MaxplanWCSP/SawtoothLP) | 2781.22 | 400 | 0.45477 | 2.13607 | 2.59084 | 15 | 412 | 1 | 89 | 0 | 13185 |
boxPushing_4_1
PARAMS
| MAX_TRIAL | MAX_TIME | Error | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|
| 34464 | 1800 | 0.01 | 4 | 0.1 | 0.01 | 0.1 |
RESULTS
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| SIM | 1801.68 | 3 | 0.250605 | 97.7405 | 97.9911 | 7 | 7 | 1 | 179 | 0 | 7387 |
| SER | 52.832 | 14 | 0 | 96.4869 | 96.4869 | 23 | 23 | 1 | 309 | 0 | 8303 |
| SIM (Maxplan/Sawtooth) | 2660.94 | 10 | 0.109513 | 98.1692 | 98.2787 | 11 | 29 | 1 | 4124 | 0 | 14896 |
| SER (Maxplan/Sawtooth) | 1807.42 | 290 | 0.01584 | 98.1692 | 98.185 | 239 | 1992 | 1 | 3682 | 0 | 14876 |
| SIM (MaxplanWCSP/SawtoothLP) | 81.6237 | 14 | -0.0119547 | 98.1692 | 98.1572 | 13 | 43 | 1 | 285 | 0 | 13257 |
boxPushing_10_2_ser(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 20 | 0.1 | 0.1 | 0.1 |
boxPushing_10_2_ser(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36025.9 | 0 | 331.237 | -102 | 229.237 | 0 | 0 | 1 | 613 | 0 | 4236 |
boxPushing_10_2_sim(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 10 | 0.1 | 0.1 | 0.1 |
boxPushing_10_2_sim(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36025.9 | 0 | 331.235 | -102 | 229.235 | 4 | 0 | 1 | 259 | 0 | 4249 |
GridSmall_10_2_ser(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 20 | 0.01 | 0.01 | 0.1 |
GridSmall_10_2_ser(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36009.4 | 2465 | 1.39427 | 5.22931 | 6.62358 | 18767 | 18767 | 1 | 32076 | 0 | 6536 |
GridSmall_10_2_sim(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 10 | 0.01 | 0.01 | 0.1 |
GridSmall_10_2_sim(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36025.5 | 9 | 1.27023 | 5.26675 | 6.53699 | 44 | 59 | 1 | 44587 | 0 | 4460 |
tiger_10_3_ser(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 20 | 0.01 | 0.01 | 0.1 |
tiger_10_3_ser(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36025.7 | 2634 | 36.5813 | 10.6306 | 47.2119 | 7413 | 7413 | 1 | 404 | 0 | 5704 |
tiger_10_3_sim(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 10 | 0.01 | 0.01 | 0.1 |
tiger_10_3_sim(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36025.8 | 0 | 1070.51 | -1010 | 60.5106 | 11 | 1 | 1 | 101 | 0 | 4212 |
Mars_10_2_ser(PARAMS)
| MAX_TRIAL | MAX_TIME | Error | Discount | Horizon | p_o | p_b | p_c |
|---|---|---|---|---|---|---|---|
| 34464 | 36000 | 0.01 | 20 | 0.1 | 0.1 | 0.1 |
Mars_10_2_ser(RESULTS)
| Time | Trials | Error | LB Value | UB Value | Total Size LB | Total Size UB | Num Nodes (oState graph) | Num Nodes (belief graph) | Num Max of JHistory | Memory |
|---|---|---|---|---|---|---|---|---|---|---|
| 36025.8 | 1 | 5.53163 | 23.0817 | 28.6133 | 27 | 26 | 1 | 850 | 0 | 4286 |
← HSVI Q-learning →