Complete result settings for Ad dataset (table columns are sortable)
The experiments are described in Rooshenas and Lowd, Discriminative Structure Learning of Arithmetic Circuits, AIStats 16
Standard Deviation | L1 Penalty | Train CLL | Validation CLL | Test CLL | Node# | Edge# | Feature# | Learning Time |
0.5 | 2 | -4.900620 | -5.599270 | -5.450508 | 253628 | 565827 | 9740 | 24595.609898 |
0.1 | 2 | -5.386751 | -6.071616 | -5.899035 | 253628 | 565827 | 9740 | 24922.696173 |
0.1 | 0.1 | -4.623933 | -5.348395 | -5.183854 | 253628 | 565827 | 9740 | 33455.159041 |
0.5 | 0.1 | -4.617104 | -5.341640 | -5.182352 | 253628 | 565827 | 9740 | 33467.003241 |
0.1 | 1 | -4.330628 | -5.099516 | -4.929483 | 253628 | 565827 | 9740 | 35988.235956 |
0.5 | 1 | -4.335505 | -5.108857 | -4.921654 | 253628 | 565827 | 9740 | 39270.217018 |
0.1 | 0.5 | -3.999095 | -4.787533 | -4.658431 | 253628 | 565827 | 9740 | 39693.478674 |
0.5 | 0.5 | -4.009455 | -4.816224 | -4.668208 | 253628 | 565827 | 9740 | 55846.351064 |