dataset,experimentnumber,method,ERT_L1_miscoverage,ERT_brier_score,ERT_logloss,ERT_underconfident_L1_miscoverage,ERT_underconfident_brier_score,ERT_underconfident_logloss,ERT_overconfident_L1_miscoverage,ERT_overconfident_brier_score,ERT_overconfident_logloss
MNIST,3,likelihood,0.14586222,0.05147351,-0.1879687,0.072908886,0.0447672,-0.12836024,0.07295336,0.0067063095,-0.05960847
MNIST,3,cumulative,0.14649779,0.06042233,-0.20945218,0.07332666,0.053797506,-0.15259303,0.07317113,0.006624818,-0.056859143
MNIST,2,likelihood,0.1497911065816879,0.0557119734585285,-0.2005841434001922,0.0758288875222206,0.0488740727305412,-0.1397248059511184,0.0739622488617897,0.0068378983996808,-0.0608593896031379
MNIST,2,cumulative,0.1425510942935943,0.0567271038889884,-0.1929759681224823,0.071320004761219,0.0503882654011249,-0.1437196433544159,0.0712311416864395,0.0063388422131538,-0.0492563061416149
MNIST,0,likelihood,0.1424533426761627,0.0509853437542915,-0.1812827289104461,0.0705599933862686,0.0444578006863594,-0.1273925602436065,0.0718933492898941,0.0065275416709482,-0.0538901910185813
MNIST,0,cumulative,0.15640889108181,0.0681213364005088,-0.2373774498701095,0.079159989953041,0.0609659068286418,-0.1731125116348266,0.0772489085793495,0.0071554230526089,-0.0642649605870246
MNIST,7,likelihood,0.1417333334684372,0.0486655607819557,-0.1812274903059005,0.0711888894438743,0.0421838983893394,-0.1222796291112899,0.0705444663763046,0.0064816577360033,-0.05894785374403
MNIST,7,cumulative,0.1509822309017181,0.0627194419503212,-0.2190546542406082,0.0753022208809852,0.0557746104896068,-0.1577455699443817,0.075680024921894,0.0069448216818273,-0.0613091066479682
MNIST,4,likelihood,0.1477600038051605,0.0520412735641002,-0.1911967247724533,0.0750577747821807,0.0453081354498863,-0.1302237808704376,0.0727022439241409,0.0067331357859075,-0.0609729886054992
MNIST,4,cumulative,0.1576177775859832,0.0692372098565101,-0.2355289012193679,0.0814311057329177,0.0622463896870613,-0.1756691336631775,0.0761866942048072,0.0069908248260617,-0.0598597824573516
MNIST,6,likelihood,0.1470533460378646,0.0527079589664936,-0.1910722255706787,0.074648879468441,0.0460347011685371,-0.1321726590394973,0.0724044665694236,0.0066732480190694,-0.0588995814323425
MNIST,6,cumulative,0.1490222215652465,0.062076486647129,-0.2103607207536697,0.0754888802766799,0.0554191246628761,-0.1572201550006866,0.0735333487391471,0.0066573619842529,-0.0531405918300151
MNIST,5,likelihood,0.1465155780315399,0.0516081824898719,-0.1885746270418167,0.0729800015687942,0.0447955913841724,-0.1286033391952514,0.0735355764627456,0.0068125920370221,-0.0599713437259197
MNIST,5,cumulative,0.1464844495058059,0.0611220076680183,-0.2067477703094482,0.0741644352674484,0.0547041296958923,-0.1550405919551849,0.0723200216889381,0.0064178751781582,-0.0517071560025215
MNIST,1,likelihood,0.1418666541576385,0.0488265193998813,-0.1768206357955932,0.0709222108125686,0.0423989631235599,-0.1221840605139732,0.0709444656968116,0.0064275562763214,-0.0546365864574909
MNIST,1,cumulative,0.1516711264848709,0.0641587674617767,-0.2206176966428756,0.07782444357872,0.057434432208538,-0.1629840433597564,0.0738466829061508,0.0067243338562548,-0.0576336495578289
MNIST,8,likelihood,0.1457022428512573,0.051355130970478,-0.187898963689804,0.0728511065244674,0.0446536652743816,-0.1282418072223663,0.0728511363267898,0.0067014633677899,-0.0596571862697601
MNIST,8,cumulative,0.1569288969039917,0.0682881474494934,-0.2368713170289993,0.0780311077833175,0.0608865618705749,-0.172508031129837,0.0788978040218353,0.0074015795253217,-0.0643633157014846
FashionMNIST,0,likelihood,0.0935435965657234,0.0145613830536603,-0.0560608617961406,0.0456948690116405,0.0109796375036239,-0.0388908274471759,0.0478487387299537,0.0035817474126815,-0.0171700604259967
FashionMNIST,0,cumulative,0.1677794903516769,0.077443704009056,-0.2742542624473572,0.0839025527238845,0.0693135261535644,-0.1960242688655853,0.0838769450783729,0.0081301750615239,-0.0782299861311912
MNIST,9,likelihood,0.1440400034189224,0.0499320849776268,-0.1844325363636016,0.0702533274888992,0.04306211322546,-0.1239229291677475,0.0737866908311843,0.0068699717521667,-0.0605096034705638
MNIST,9,cumulative,0.1423422396183014,0.0566375777125358,-0.1935073435306549,0.0720377787947654,0.0503912344574928,-0.1433252394199371,0.0703044682741165,0.006246343255043,-0.0501821115612983
FashionMNIST,1,likelihood,0.1007025688886642,0.018410075455904,-0.0803355872631073,0.0498128160834312,0.0142289195209741,-0.0481943897902965,0.0508897602558136,0.0041811554692685,-0.0321412086486816
FashionMNIST,1,cumulative,0.1579179465770721,0.0671742931008338,-0.2390991896390915,0.0774461552500724,0.059530384838581,-0.1691414266824722,0.0804718136787414,0.0076439096592366,-0.0699577778577804
FashionMNIST,2,likelihood,0.102400004863739,0.0182023830711841,-0.0711004585027694,0.0513410195708274,0.0141766117885708,-0.0486682578921318,0.0510589964687824,0.0040257708169519,-0.0224322080612182
FashionMNIST,2,cumulative,0.165923073887825,0.074704423546791,-0.2666301727294922,0.0832307711243629,0.0666836649179458,-0.1889974772930145,0.0826923251152038,0.0080207539722323,-0.0776327103376388
FashionMNIST,3,likelihood,0.0973948687314987,0.0143124042078852,-0.0667832344770431,0.0479794815182685,0.0102731734514236,-0.03778987377882,0.0494154021143913,0.0040392307564616,-0.0289933625608682
FashionMNIST,3,cumulative,0.1676102578639984,0.0765616297721862,-0.2698097825050354,0.0847923010587692,0.0685791596770286,-0.194826453924179,0.0828179717063903,0.0079824700951576,-0.0749833434820175
FashionMNIST,4,likelihood,0.1063794940710067,0.0190451387315988,-0.0772144570946693,0.0562538392841815,0.0150750633329153,-0.05195839330554,0.0501256585121154,0.0039700763300061,-0.0252560377120971
FashionMNIST,4,cumulative,0.1587846279144287,0.0687064006924629,-0.2406060248613357,0.0783923119306564,0.0611376352608203,-0.1719611287117004,0.0803923234343528,0.0075687705539166,-0.0686449036002159
FashionMNIST,5,likelihood,0.0961179584264755,0.0152944084256887,-0.0654350742697715,0.046866662800312,0.0113193644210696,-0.040074236690998,0.0492512956261634,0.0039750398136675,-0.0253608468919992
FashionMNIST,5,cumulative,0.1683025658130645,0.0762683898210525,-0.2707467675209045,0.0856256335973739,0.0682495236396789,-0.19313645362854,0.0826769396662712,0.0080188680440187,-0.0776102989912033
FashionMNIST,6,likelihood,0.0952051430940628,0.0150144500657916,-0.0670882165431976,0.0459102503955364,0.0111024426296353,-0.0390143319964408,0.049294888973236,0.0039120065048336,-0.0280738826841115
FashionMNIST,6,cumulative,0.1686718016862869,0.0764194428920745,-0.273015558719635,0.0857846140861511,0.0683731585741043,-0.1931429058313369,0.0828871950507164,0.0080462917685508,-0.0798726826906204
FashionMNIST,8,likelihood,0.0948666781187057,0.0143194962292909,-0.0663348734378814,0.0444589704275131,0.0102298213168978,-0.0367770195007324,0.0504077151417732,0.0040896697901189,-0.0295578595250844
FashionMNIST,8,cumulative,0.1635436117649078,0.0730136036872863,-0.2574922442436218,0.0828102603554725,0.0652521625161171,-0.1849315315485,0.0807333514094352,0.0077614425681531,-0.0725607424974441
FashionMNIST,7,likelihood,0.1021641045808792,0.0163998790085315,-0.0743558779358863,0.0521461479365825,0.0122840274125337,-0.0439800918102264,0.0500179640948772,0.0041158497333526,-0.0303757842630147
FashionMNIST,7,cumulative,0.1666615307331085,0.0745042562484741,-0.2651689052581787,0.0840102434158325,0.0665068328380584,-0.1881427466869354,0.0826513022184372,0.0079974178224802,-0.0770261660218238
FashionMNIST,9,likelihood,0.0908923149108886,0.0139142023399472,-0.0564302578568458,0.0416640974581241,0.0102443965151906,-0.0359168127179145,0.0492282211780548,0.0036698072217404,-0.0205134209245443
FashionMNIST,9,cumulative,0.1599384695291519,0.0683872550725936,-0.2442404776811599,0.0801358968019485,0.0607761964201927,-0.1726949512958526,0.0798025801777839,0.0076110465452075,-0.0715455114841461
CIFAR10,4,likelihood,0.019560004,-0.0036628903,0.025515944,0.0109299915,-0.0026677684,0.006624603,0.008630015,-0.0009951219,0.018891329
CIFAR10,4,cumulative,0.07773001,0.012350604,-0.029212058,0.039739996,0.010439749,-0.03546956,0.037990015,0.0019108519,0.006257498
CIFAR10,0,likelihood,0.0138500053435564,-0.0034597397316247,0.0241651348769664,0.0072749927639961,-0.0023687824141234,0.0065215467475354,0.0065750144422054,-0.0010909556876868,0.0176435951143503
CIFAR10,0,cumulative,0.0724900066852569,0.011398859322071,-0.0154948532581329,0.0384199917316436,0.0101732267066836,-0.0345428101718425,0.0340700186789035,0.0012256338959559,0.0190479345619678
CIFAR10,2,likelihood,0.0186600051820278,-0.0033202380873262,0.0312386509031057,0.0081299915909767,-0.0022520557977259,0.0059740305878221,0.0105300145223736,-0.0010681882267817,0.0252646096050739
CIFAR10,2,cumulative,0.0670300051569938,0.0058912397362291,-0.0082942126318812,0.033514991402626,0.0045413943007588,-0.0196180455386638,0.0335150137543678,0.0013498484622687,0.0113238040357828
CIFAR10,6,likelihood,0.014210008084774,-0.0034432872198522,0.0320811159908771,0.0056549930013716,-0.0020136714447289,0.0057010650634765,0.0085550164803862,-0.0014296189183369,0.0263800509274005
CIFAR10,6,cumulative,0.076210007071495,0.0113353822380304,-0.0210977979004383,0.0358549915254116,0.0092533379793167,-0.0315970107913017,0.0403550192713737,0.0020820409990847,0.0104991914704442
CIFAR10,3,likelihood,0.0168800074607133,-0.0033836036454886,0.0329720079898834,0.0083899926394224,-0.002040462102741,0.0053588687442243,0.0084900157526135,-0.0013431400293484,0.0276131220161914
CIFAR10,3,cumulative,0.0738100111484527,0.0110116750001907,-0.0226106587797403,0.0388049930334091,0.0094347521662712,-0.0323136039078235,0.0350050181150436,0.0015769272577017,0.0097029209136962
CIFAR10,1,likelihood,0.0145000042393803,-0.0039037480019032,0.0241370853036642,0.0063499906100332,-0.0030231238342821,0.0083357095718383,0.0081500131636857,-0.0008806184050627,0.0158013347536325
CIFAR10,1,cumulative,0.0706800073385238,0.0097640976309776,-0.0135399159044027,0.0339149944484233,0.0082609932869672,-0.0286388806998729,0.036765020340681,0.0015031039947643,0.0150989834219217
CIFAR10,7,likelihood,0.0154800023883581,-0.0043416456319391,0.0243897084146738,0.0069399909116327,-0.003558540251106,0.0100568952038884,0.0085400138050317,-0.0007831052062101,0.0143327889963984
CIFAR10,7,cumulative,0.0709700137376785,0.010905859991908,-0.0144312502816319,0.0365849956870079,0.0096931103616952,-0.0328267142176628,0.0343850180506706,0.0012127548689022,0.0183954592794179
CIFAR10,5,likelihood,0.017890004441142,-0.0031647160649299,0.0213245749473571,0.0070449910126626,-0.0023996471427381,0.0067881224676966,0.0108450148254632,-0.0007650658371858,0.014536440372467
CIFAR10,5,cumulative,0.070980004966259,0.0091474931687116,-0.0082960249856114,0.0381399989128112,0.0079536959528923,-0.0290933661162853,0.0328400172293186,0.0011938035022467,0.0207973234355449
CIFAR10,9,likelihood,0.0166700053960084,-0.0032874182797968,0.0255378782749176,0.0080849910154938,-0.0023436560295522,0.0064625204540789,0.0085850143805146,-0.0009437635308131,0.0190753638744354
CIFAR10,9,cumulative,0.0629400163888931,0.006684198975563,-0.010891580954194,0.0303949955850839,0.0055299522355198,-0.0209290869534015,0.032545018941164,0.0011542498832568,0.0100374761968851
CIFAR10,8,likelihood,0.0135500058531761,-0.0046122609637677,0.0419189520180225,0.0106499921530485,-0.0025330453645437,0.0058393776416778,0.0029000148642808,-0.0020792125724256,0.0360795632004737
CIFAR10,8,cumulative,0.0780100002884864,0.0099108116701245,-0.0290978197008371,0.0427549928426742,0.0080636236816644,-0.0301345698535442,0.0352550148963928,0.0018471836810931,0.0010367572540417
CIFAR100,0,likelihood,0.0020700153,-0.035049707,0.4215755,0.003609997,-0.027321484,0.085097745,-0.0015399812,-0.0077282228,0.3364777
CIFAR100,0,cumulative,0.03462002,-0.005746712,0.22180481,0.016459998,-0.0015428156,0.00675382,0.018160021,-0.0042038932,0.21505101
CIFAR100,7,likelihood,0.0091400165110826,-0.02791428565979,0.3929481506347656,0.0043199970386922,-0.0212741084396839,0.0657429844141006,0.0048200194723904,-0.0066401697695255,0.3272052407264709
CIFAR100,7,cumulative,0.0464100129902362,-0.0071611255407333,0.1555384695529937,0.0223299972712993,-0.0046221585944294,0.0140380980446934,0.0240800194442272,-0.0025389655493199,0.1415003985166549
CIFAR100,6,likelihood,0.0102400165051221,-0.0343610160052776,0.4233700037002563,0.004069996997714,-0.0279235653579235,0.0913891047239303,0.0061700199730694,-0.0064374506473541,0.3319808542728424
CIFAR100,6,cumulative,0.0525700151920318,-0.0062531665898859,0.1540416330099105,0.0261599961668252,-0.0042467219755053,0.0150401834398508,0.0264100190252065,-0.0020064427517354,0.1390014290809631
CIFAR100,2,likelihood,0.0111700156703591,-0.034333050251007,0.3927448987960815,0.0034099970944225,-0.0281434953212738,0.0899019241333007,0.0077600199729204,-0.0061895474791526,0.3028429746627807
CIFAR100,2,cumulative,0.0398600175976753,-0.0125298965722322,0.187033399939537,0.0172299984842538,-0.0097387535497546,0.0340038426220417,0.0226300172507762,-0.0027911409270018,0.1530295312404632
CIFAR100,1,likelihood,0.0027500146534293,-0.0417894981801509,0.4246869683265686,0.0048999963328242,-0.0345001108944416,0.1148479580879211,-0.0021499814465641,-0.0072893910109996,0.3098390698432922
CIFAR100,1,cumulative,0.0397300198674202,-0.0074731530621647,0.1784892976284027,0.0197149980813264,-0.0040602684020996,0.0130909141153097,0.0200150199234485,-0.0034128874540328,0.1653984040021896
CIFAR100,3,likelihood,0.0041000172495841,-0.0273350402712821,0.4296560883522033,0.0021999976597726,-0.0198013372719287,0.0605565011501312,0.0019000204047188,-0.0075337067246437,0.3690996766090393
CIFAR100,3,cumulative,0.0420300178229808,-0.0064946324564516,0.1889880448579788,0.0214149989187717,-0.0030114755500108,0.0118986191228032,0.0206150207668542,-0.0034831569064408,0.1770894229412078
CIFAR100,4,likelihood,0.0078300144523382,-0.0359487272799015,0.4059362411499023,0.0053899958729743,-0.0292496457695961,0.0913512557744979,0.0024400190450251,-0.0066990815103054,0.3145850300788879
CIFAR100,4,cumulative,0.0376500152051448,-0.0066740587353706,0.227185845375061,0.0187499988824129,-0.0026317059528082,0.0126562835648655,0.0189000219106674,-0.0040423469617962,0.2145295888185501
CIFAR100,8,likelihood,0.0052500152960419,-0.0400752462446689,0.4214964210987091,0.0033999965526163,-0.0331853032112121,0.1039576530456543,0.0018500194419175,-0.0068899542093276,0.3175387680530548
CIFAR100,8,cumulative,0.0377100184559822,-0.0086442483589053,0.2013329565525055,0.0205549970269203,-0.0047645047307014,0.020345514640212,0.0171550195664167,-0.0038797452580183,0.1809874475002288
CIFAR100,9,likelihood,0.0085700172930955,-0.0228415876626968,0.3474809527397156,0.0027349973097443,-0.0164350159466266,0.0454254150390625,0.0058350199833512,-0.0064065717160701,0.3020554780960083
CIFAR100,9,cumulative,0.0363900177180767,-0.0104629416018724,0.1998827010393142,0.0185949988663196,-0.0066978172399103,0.0225508864969015,0.0177950207144022,-0.0037651241291314,0.1773318201303482
CIFAR100,5,likelihood,0.0057800151407718,-0.0406671576201915,0.4265289008617401,0.0031899963505566,-0.0337297469377517,0.1040275245904922,0.0025900192558765,-0.0069373995065689,0.3225014209747314
CIFAR100,5,cumulative,0.0410200133919715,-0.0062809498049318,0.1912094056606292,0.0243349988013505,-0.0021740137599408,0.0098579106852412,0.0166850201785564,-0.0041069374419748,0.1813515424728393