Complete result settings for Accidents 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	-26.898108	-26.845574	-27.108686	848586	1942166	2324	10474.272667
10	0.5	2	-26.846354	-26.813322	-27.064223	806399	1845762	2468	10990.166240
10	0.1	0.1	-26.897186	-26.869559	-27.145097	876455	1985915	2584	11418.163174
10	0.1	1	-26.823333	-26.799051	-27.054149	841055	1938354	2596	12020.128662
10	0.5	1	-26.532924	-26.541107	-26.745268	826889	1907546	2430	13427.401724
10	0.5	0.1	-26.724506	-26.716752	-26.974291	844719	1944001	2768	15664.155685
5	0.5	2	-26.433275	-26.455820	-26.661766	837626	1939594	3336	18263.365546
5	0.1	1	-26.635540	-26.620367	-26.826297	810668	1853028	3522	24051.010689
5	0.1	2	-26.734733	-26.715399	-26.939615	787253	1799486	3092	21985.071760
5	0.5	1	-26.435426	-26.421289	-26.666847	860243	1971557	3342	25950.858869
5	0.5	0.1	-26.456208	-26.465928	-26.720061	863965	1956320	4002	33831.061896
5	0.1	0.1	-26.688105	-26.714681	-26.928418	828126	1896012	4120	33046.192214
2	0.1	2	-26.486440	-26.541165	-26.763255	804037	1842765	4440	43860.915126
2	0.1	1	-26.474492	-26.531669	-26.754195	841284	1912415	4814	45932.934132
2	0.5	2	-26.287916	-26.328886	-26.571006	840961	1916274	4780	55238.842418
2	0.5	1	-26.230223	-26.238895	-26.558994	838835	1908961	5182	56192.081504
2	0.1	0.1	-26.338022	-26.374233	-26.648498	834779	1900359	6342	67838.374998
2	0.5	0.1	-26.332793	-26.407210	-26.679666	618018	1375714	6214	76698.253093