Complete result settings for Netflix 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 |
0.1 | 0.1 | -10.828564 | -10.942223 | -10.998452 | 578278 | 1334757 | 2574 | 57498.030970 |
0.5 | 0.1 | -10.829723 | -10.944287 | -10.999903 | 578278 | 1334757 | 2574 | 61369.665392 |
0.1 | 0.5 | -10.828353 | -10.941387 | -10.996824 | 578278 | 1334757 | 2574 | 64094.906092 |
0.1 | 1 | -10.828389 | -10.940099 | -10.996761 | 578278 | 1334757 | 2574 | 66534.791174 |
0.5 | 0.5 | -10.830280 | -10.942656 | -10.998823 | 578278 | 1334757 | 2574 | 76999.299326 |
0.5 | 1 | -10.828384 | -10.940141 | -10.996794 | 578278 | 1334757 | 2574 | 79254.226525 |