Complete result settings for MSNBC 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	-2.916540	-2.922757	-2.925277	213127	431453	10034	5630.558024
0.5	1	-2.916539	-2.922754	-2.925276	213127	431453	10034	6350.365596
0.5	0.5	-2.916538	-2.922753	-2.925276	213127	431453	10034	7908.249763
0.5	0.1	-2.916537	-2.922753	-2.925275	213127	431453	10034	8652.169670
0.1	2	-2.916540	-2.922757	-2.925277	213127	431453	10034	8698.376645
0.1	0.1	-2.916537	-2.922753	-2.925275	213127	431453	10034	7873.293077
0.1	0.5	-2.916538	-2.922753	-2.925276	213127	431453	10034	7665.337691
0.1	1	-2.916539	-2.922754	-2.925276	213127	431453	10034	8189.576994