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

Split Penalty	Standard Deviation	L1 Penalty	Train LL	Validation LL	Test LL	Node#	Edge#	Feature#	Learning Time
10	0.5	0.1	-56.854292	-57.062951	-57.125649	802522	1848704	1784	17302.324646
10	0.1	2	-56.846825	-57.052846	-57.091583	792968	1840181	1744	19360.843704
10	0.1	1	-56.932084	-57.125190	-57.193599	773962	1781918	1692	20368.406530
10	0.5	1	-56.930897	-57.132945	-57.216786	787309	1817901	1732	22127.914045
10	0.5	2	-56.868302	-57.067034	-57.144139	793846	1842812	1720	24077.124720
10	0.1	0.1	-56.858847	-57.064823	-57.138484	756689	1764895	1782	23948.702243
5	0.5	0.1	-56.661038	-56.963333	-57.028078	837144	1944537	2548	38750.589015
5	0.1	0.1	-56.691312	-56.999665	-57.037474	733245	1693402	2498	43835.665965
5	0.5	1	-56.653308	-56.960187	-57.008915	843911	1943810	2568	44105.533939
5	0.5	2	-56.655097	-56.908882	-56.997547	846983	1948255	2468	45152.509774
5	0.1	2	-56.655304	-56.917788	-56.989865	699094	1592676	2504	45890.523579
5	0.1	1	-56.627625	-56.884989	-56.964193	578278	1334757	2574	48129.804156
2	0.5	2	-56.518105	-56.904328	-56.969063	572943	1306458	3996	76900.879288
2	0.1	0.1	-56.544090	-56.947033	-57.014190	605785	1381211	4254	77299.933623
2	0.1	2	-56.532536	-56.889619	-56.927805	625946	1436287	3718	77377.246869
2	0.1	1	-56.482468	-56.888607	-56.930988	615020	1407255	3928	77323.810993
2	0.5	0.1	-56.488221	-56.891118	-56.987228	606432	1373407	4206	77182.107535
2	0.5	1	-56.502101	-56.920095	-56.969414	631828	1444974	4098	77279.854675