Complete result settings for DNA 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	0.1	-11.217556	-12.222106	-12.159474	295508	677391	2136	2561.526588
0.5	2	-11.309448	-12.148761	-12.099290	295508	677391	2136	3745.692567
0.1	2	-11.408338	-12.199633	-12.152999	295508	677391	2136	7315.034944
0.5	0.5	-11.160217	-12.200401	-12.145984	295508	677391	2136	8040.388674
0.1	1	-11.204428	-12.165830	-12.113104	295508	677391	2136	8250.197778
0.1	0.1	-11.264345	-12.206294	-12.152499	295508	677391	2136	11076.557107
0.5	1	-11.207080	-12.158595	-12.109946	295508	677391	2136	12238.949396
0.1	0.5	-11.158673	-12.208426	-12.144362	295508	677391	2136	12807.118021
0.5	0.1	-11.217556	-12.222106	-12.159474	295508	677391	2136	2729.359073
0.1	0.1	-11.264345	-12.206294	-12.152499	295508	677391	2136	15057.976838