Complete result settings for EachMovie 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	0.1	2	-56.510887	-61.235002	-56.612049	680283	1594994	3814	28728.285636
10	0.1	1	-56.448012	-61.060094	-56.452288	730181	1719540	3944	41451.025485
10	0.1	0.1	-56.361598	-61.012617	-56.366510	839343	1964749	4346	41536.166542
10	0.5	1	-55.757892	-60.512489	-56.041407	845406	1975229	4216	50154.137411
10	0.5	2	-55.801917	-60.593043	-56.008359	788825	1854763	4114	50100.489567
10	0.5	0.1	-55.622512	-60.663332	-56.079529	821233	1904117	4574	59431.216081
2	0.5	0.1	-55.610131	-60.628414	-56.106079	117667	249778	7534	75116.839504
2	0.5	2	-54.769469	-59.633044	-55.354190	168058	371495	6324	75206.799827
2	0.5	1	-54.499685	-59.599975	-55.120183	182557	405995	6710	75197.934175
2	0.1	1	-55.161493	-59.898337	-55.349116	168797	372264	6582	75231.298103
5	0.5	0.1	-55.650707	-60.648951	-55.968633	184214	410941	5928	75404.741736
5	0.1	0.1	-55.787357	-60.471207	-55.801355	396903	910860	5838	75452.273510
2	0.1	0.1	-55.896170	-60.718513	-56.048038	133424	285965	7692	75653.804872
2	0.1	2	-55.080796	-59.810418	-55.354352	266645	605496	6318	75715.451500
5	0.5	1	-54.909483	-59.752528	-55.278418	376542	870167	5326	75812.673721
5	0.1	2	-55.370536	-60.106600	-55.594106	776132	1803851	4846	75987.021216
5	0.5	2	-54.880449	-59.687840	-55.396053	767627	1785756	5132	76108.767708
5	0.1	1	-55.493917	-60.268939	-55.644965	809209	1878812	4986	76306.983574