Complete result settings for WebKB 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	0.5	-26.489713	-29.558751	-29.669784	596676	1376735	7684	23564.816603
0.1	0.5	-26.489227	-29.556381	-29.663841	596676	1376735	7684	24944.460865
0.5	1	-26.512972	-29.498477	-29.592999	596676	1376735	7684	41729.792106
0.5	2	-26.568917	-29.445816	-29.524813	596676	1376735	7684	48074.889505
0.1	2	-26.563574	-29.448977	-29.522169	596676	1376735	7684	57405.720003
0.1	1	-26.506246	-29.510225	-29.602314	596676	1376735	7684	74438.917563
0.5	0.1	-26.477531	-29.622888	-29.731081	596676	1376735	7684	31551.577431
0.1	0.1	-26.477916	-29.621724	-29.732672	596676	1376735	7684	40684.514013