OpenBench

OpenBench Testing Framework

Author	Menkib
Creation	1778169649
Source	https://github.com/Menkib64/lc0
Sha	14163e5c87046657c33b2d6e767ec3087b7cd53b
Branch	uncertainty_weighting_logistic
Bench	1
Network	tf13-375
Options	Threads=2 Hash=1
Time	20.0+0.20
Opening Book	UHO_Lichess_4852_v1.epd
Upload PGNs	FALSE
Syzygy WDL	DISABLED
Syzygy ADJ.	OPTIONAL
Win ADJ.	movecount=3 score=400
Draw ADJ.	movenumber=40 movecount=8 score=10
Scaling	DEV 6250 NPS
Alpha	0.75
Gamma	0.19
A-Ratio	0.1
Reporting	BATCHED
Distribution	SINGLE
Pairs Per Point	8

uncertainty_weighting_lo... vs uncertainty_weighting_lo...

Approve Restart Delete

UncertaintyWeightingMidPoint, float, 0.046, 0.0, 1.0, 0.02, 0.02
UncertaintyWeightingMinusExponent, float, 117.2, 0.0, 100000.0, 0.5, 0.5
UncertaintyWeightingScale, float, 0.404, 0.0, 10.0, 0.02, 0.02
UncertaintyWeightingBase, float, 0.673, 0.0, 10.0, 0.02, 0.02
UncertaintyWeightingSkewRoot, float, 0.79, 0.01, 10000.0, 0.1, 0.1

Copy SPSA Inputs

UncertaintyWeightingMidPoint, 0.07850672750932469
UncertaintyWeightingMinusExponent, 126.53922512827481
UncertaintyWeightingScale, 0.3626439317441151
UncertaintyWeightingBase, 0.7149950324835604
UncertaintyWeightingSkewRoot, 0.5404352469785321

Copy SPSA Outputs

Tuning 5 Parameters
2830/25000 Iterations
45280/400000 Games Played

Name	Curr	Start	Min	Max	C	C_end	R	R_end
UncertaintyWeightingMidPoint	0.0785	0.0460	0.0000	1.0000	0.0303	0.0200	0.0299	0.0200
UncertaintyWeightingMinusExponent	126.5392	117.2000	0.0000	100000.0000	0.7563	0.5000	0.7480	0.5000
UncertaintyWeightingScale	0.3626	0.4040	0.0000	10.0000	0.0303	0.0200	0.0299	0.0200
UncertaintyWeightingBase	0.7150	0.6730	0.0000	10.0000	0.0303	0.0200	0.0299	0.0200
UncertaintyWeightingSkewRoot	0.5404	0.7900	0.0100	10000.0000	0.1513	0.1000	0.1496	0.1000

#	User	Updated	Games	Won	Lost	Drawn	Crash
238	Ploot	1778229544	886	241	240	405	0
413	Kovax	1778229613	798	220	208	370	0
414	Kovax	1778229586	782	171	190	421	0
416	Kovax	1778229644	784	177	186	421	0
1271	Kovax	1778229551	968	239	223	506	0
1259	Kovax	1778229598	940	249	251	440	0
1306	Kovax	1778229577	962	246	237	479	0
1268	Kovax	1778229619	964	241	233	490	0
1270	Kovax	1778229515	926	218	215	493	0
1260	Kovax	1778229577	956	227	239	490	0
1275	Kovax	1778229567	940	235	248	457	0
1261	Kovax	1778229615	926	224	216	486	0
1263	Kovax	1778229605	940	242	240	458	1
1262	Kovax	1778229649	938	240	212	486	0
1277	Kovax	1778229649	942	229	221	492	0
1253	Kovax	1778229649	938	208	241	489	0
1267	Kovax	1778229581	944	229	243	472	0
1252	Kovax	1778229568	938	242	228	468	0
1258	Kovax	1778229610	958	243	255	460	2
1266	Kovax	1778229618	950	229	253	468	0
1276	Kovax	1778229482	944	232	224	488	0
1269	Kovax	1778229552	932	243	235	454	0
1272	Kovax	1778229579	914	216	223	475	0
1265	Kovax	1778229610	954	233	233	488	0
1274	Kovax	1778229585	912	224	209	479	0
1257	Kovax	1778229563	920	209	238	473	0
1264	Kovax	1778229648	916	217	218	481	0
1273	Kovax	1778229475	944	209	226	509	0
1283	Kovax	1778229453	928	216	232	480	0
1282	Kovax	1778229598	936	237	230	469	0
1279	Kovax	1778229644	936	227	224	485	0
1280	Kovax	1778229594	948	236	229	483	0
1256	Kovax	1778229582	952	232	244	476	0
1278	Kovax	1778229582	950	230	237	483	0
1281	Kovax	1778229641	938	223	215	500	0
444	Kovax	1778183705	142	33	33	76	0
1310	Kovax	1778229607	902	204	237	461	0
1307	Kovax	1778229607	884	198	221	465	0
1312	Kovax	1778229628	914	229	217	468	0
1313	Kovax	1778229641	916	239	209	468	0
1309	Kovax	1778229528	924	234	223	467	0
1308	Kovax	1778229613	914	223	217	474	0
443	Kovax	1778183699	106	29	27	50	0
415	Kovax	1778229512	752	181	187	384	0
239	Ploot	1778229645	828	216	202	410	0
1255	Kovax	1778229558	918	211	234	473	0
237	Ploot	1778229505	830	212	207	411	0
441	Kovax	1778183737	108	28	26	54	0
1311	Kovax	1778229628	920	223	242	455	0
442	Kovax	1778183757	106	25	21	60	0
236	Ploot	1778229623	820	192	199	429	0
1314	Kovax	1778229561	478	107	126	245	0
1315	Kovax	1778229601	466	117	120	229	0
1316	Kovax	1778229603	474	114	106	254	0
1317	Kovax	1778229502	474	128	126	220	0