The Configurable Bottleneck: A Cost-Constrained Capability Map for a Deterministic Edge Decision Token

{"The":[0,174,347,620,703,793,858,904,946,1070,1127],"Configurable":[1],"Bottleneck:":[2],"A":[3,39,526,982,1138],"Cost-Constrained":[4],"Capability":[5],"Map":[6],"for":[7,228,233,594,679,1053],"a":[8,66,94,107,115,119,124,128,132,136,143,149,192,203,219,244,274,292,304,325,351,364,378,386,406,431,469,560,568,576,600,615,646,682,710,740,760,804,811,829,883,888,899,956,974,997,1039,1077,1166,1202,1239,1257,1280,1297],"Deterministic":[9],"Edge":[10],"Decision":[11],"Token":[12],"Randolph":[13],"James":[14],"Ferlic,":[15],"M.D.":[16],"and":[17,127,142,170,179,202,231,253,291,362,373,393,445,485,488,504,541,564,583,599,640,643,660,739,752,790,842,848,898,916,954,966,969,1106,1148,1207,1217,1224,1256,1316,1375,1382,1411,1423,1426],"Kimberly":[18],"Kate":[19],"Ferlic":[20],"(Fieldstone":[21],"Analytics,":[22],"LLC,":[23],"Austin,":[24],"TX,":[25],"USA)":[26],"Preprint":[27],"·":[28,32,35,857,938,990,1069,1126,1197,1269,1324,1387],"Zenodo":[29],"DOI:":[30],"10.5281/zenodo.22884023":[31],"CC-BY":[33],"4.0":[34],"Community:":[36],"spiral-domain-encoder-campaign":[37],"Abstract":[38],"frozen,":[40],"deterministic,":[41,806],"class-discriminant":[42],"single-token":[43],"encoder":[44],"(features":[45],"→":[46,51,58,1152,1156,1160,1165],"Fisher-discriminant":[47],"⊕":[48],"PCA":[49],"subspace":[50],"k-means":[52],"codebook":[53,138,422,433,507,1028,1041],"of":[54,68,282,295,341,597,617,675,831,846,1018,1141],"≤":[55,247,926],"256":[56],"cells":[57],"nearest-centroid":[59],"decision)":[60],"has":[61],"been":[62],"characterized":[63],"piecemeal":[64],"across":[65,160,401,696,786,1311],"program":[67],"papers":[69],"—":[70,81,114,154,191,237,242,336,446,450,478,487,565,567,641,653,659,753,755,819,868,910,922,995,1085,1132,1172,1289,1326,1389,1415],"its":[71,74,77,79,391,447,987,1024],"deployment":[72,95,387,794,983],"behavior,":[73],"accuracy":[75,185,366,372,392,442,537,546,607,866,958,1061],"tax,":[76,443],"label-efficiency,":[78,425],"privacy":[80,383,980,1208],"but":[82,1086],"never":[83,254,699],"as":[84,243,732,853,923,1249],"one":[85,89,279,337,512,629,750,943,1189,1267],"configurable":[86],"platform,":[87],"on":[88,91,184,196,259,264,473,502,509,611,645,669,815,892,1096,1099,1121,1146],"protocol,":[90,1402],"the":[92,157,182,186,197,209,234,256,260,265,268,283,300,332,344,354,358,368,402,414,420,437,440,456,460,497,505,520,531,536,544,556,585,595,618,654,666,670,688,691,708,726,733,766,770,776,787,799,816,843,861,865,871,875,893,932,936,950,1003,1008,1019,1026,1046,1055,1059,1064,1081,1090,1102,1108,1116,1133,1179,1185,1192,1200,1250,1272,1286,1291,1301,1327,1338,1343,1348,1353,1359,1365,1371,1378,1384,1390,1396,1403,1424,1430],"axes":[93],"engineer":[96],"actually":[97,713,1205],"trades.":[98],"This":[99,464,827],"is":[100,176,215,251,267,349,357,413,423,434,462,466,500,627,657,665,705,758,773,798,828,851,863,870,907,949,1007,1030,1042,1067,1074,1136,1278,1293],"that":[101,229,250,605,685,779,796,810,930,1304],"unified,":[102],"pre-registered,":[103],"multi-axis,":[104],"multi-domain":[105],"characterization:":[106],"base":[108,159],"token":[109,121,187,210,356,458,552,570,704,728,862,948,1092,1118,1168,1183],"(M0)":[110],"plus":[111],"seven":[112,369,474,941,960,1412],"dials":[113,370,521,621,663,847,962,1128],"soft-vote":[116,1047],"readout":[117,1312],"(D1),":[118],"channel-partition":[120],"ladder":[122,438,1056,1082,1366,1418],"(D2),":[123],"spiral":[125,533,1170,1416],"(D3)":[126],"diverse-model":[129,411,498,557,890,1004],"(D4)":[130,1006],"co-channel,":[131],"noise-robust":[133],"reader":[134],"(D5),":[135],"foundation":[137,421,506,1027],"with":[139,405,511,548,886,996,1329],"shrinkage":[140],"(D6),":[141],"cheap":[144,193,416,601,889,1010],"quantized-sketch":[145,602],"co-channel":[146,412,499,534,603,742,1005,1171,1240,1259],"(D7),":[147],"each":[148,376,781,972,1306],"previously-filed":[150],"or":[151,382,1241],"-published":[152],"method":[153],"measured":[155,908],"against":[156,273,331],"same":[158,198,894],"ECG,":[161],"industrial":[162],"vibration,":[163],"surgical-robot":[164],"kinematics,":[165],"wearable":[166],"EMG,":[167],"financial":[168],"volatility,":[169],"continuous":[171],"glucose":[172],"data.":[173],"headline":[175,449,1073],"deliberately":[177],"honest":[178,859],"it":[180,217,277,313,610,835,869,1299],"reframes":[181],"object:":[183],"does":[188],"not":[189,623,864,1130,1195],"win":[190],"random-forest":[194],"ensemble":[195,230],"features":[199,895],"matches":[200,604,896],"token-plus-co-channel,":[201],"gradient-boosted":[204,900],"model":[205,236,311,901],"wins":[206,902],"outright.":[207,903],"What":[208,1320],"wins,":[211],"by":[212,468,1191,1238],"direct":[213],"measurement,":[214],"cost:":[216],"reaches":[218],"decision":[220,360,741,952,1258],"in":[221,1333,1394],"3–5":[222,911],"microseconds":[223],"per":[224,553,820,822,824],"window":[225,285],"versus":[226,913],"40–190":[227,914],"20–300":[232,917],"full":[235],"10":[238],"to":[239,321,338,694,707,1199,1212],"60×":[240],"cheaper":[241,921],"fully":[245],"enumerable":[246,925],"256-entry":[248,927],"table":[249,929],"auditable":[252,928],"puts":[255,299],"raw":[257,284,648,749,933,1123,1266],"signal":[258,934],"wire.":[261,937],"That":[262],"silence":[263],"wire":[266],"token's":[269,905],"largest":[270],"cost":[271,872,909,1223],"advantage:":[272],"full-data-stream-retention":[275],"architecture":[276],"emits":[278],"byte":[280],"instead":[281],"(an":[286,736,1253],"exact":[287],"180×–24,000×":[288],"data":[289],"reduction),":[290],"pre-registered":[293,470,642,1139],"Monte-Carlo":[294,1425],"radio-dominated":[296],"device":[297,320],"energy":[298,1428],"transmission-energy":[301],"savings":[302],"at":[303,377,427,435,489,559,575,614,680,719,973,1014,1022,1035,1043,1220],"median":[305],"≈":[306,315,322,328],"350×":[307],"(a":[308,515,588,672],"Markov":[309,1427],"duty-cycle":[310],"brackets":[312],"from":[314,725],"14×":[316],"vs":[317,324],"an":[318,747,924,1264],"event-triggered":[319],"870×":[323],"continuous-streaming":[326],"one;":[327],"900×":[329],"even":[330],"strongest":[333],"decision-preserving":[334],"codec)":[335],"two":[339,1407],"orders":[340],"magnitude":[342],"beyond":[343],"compute":[345,757,976,1277],"figure.":[346],"platform":[348],"therefore":[350],"cost-versus-capability":[352],"frontier:":[353],"bare":[355,457,947,1091,1153],"cheapest":[359,777,951,1302],"floor":[361,953],"concedes":[363,955],"real":[365,448,957,1072],"tax;":[367,959],"buy":[371,963],"robustness":[374,613,1211],"back,":[375,971],"stated":[379,561,975],"compute,":[380],"bit,":[381],"cost,":[384],"so":[385,769,1290],"spends":[388,984],"only":[389,539,985],"what":[390,986],"threat":[394],"budget":[395,988],"require.":[396],"Each":[397,991],"dial's":[398,992],"benefit":[399,1021],"generalizes":[400,994],"six":[403,1001],"applications":[404],"clear":[407],"per-dial":[408,998],"pattern":[409],"(the":[410,1176,1400],"universal":[415,424,1009,1031],"accuracy/noise":[417,1011],"dial,":[418],"+0.05–0.19;":[419],"near-ceiling":[426],"eight":[428,1036,1330],"labels":[429,1037],"where":[430,455,519,1038,1115],"from-scratch":[432],"chance;":[436],"recovers":[439,535,1058],"multichannel":[441,649,1060,1124],"+0.12,":[444],"holds":[451,1104],"under":[452,637,1087,1296],"sensor":[453,712,1088,1204],"dropout":[454,1089,1159],"collapses;":[459],"noise-reader":[461,1065],"additive-noise-specific).":[463],"generalization":[465,1355],"confirmed":[467],"breadth":[471,1361,1409],"battery":[472],"further":[475,529],"physically-distinct":[476],"modalities":[477],"inertial,":[479],"chemical,":[480],"acoustic/speech,":[481],"gait,":[482],"surgical-robotics,":[483],"sleep/respiratory,":[484],"EEG":[486],"population":[490,734,1251],"scale":[491],"(21,699":[492],"ECG":[493],"records,":[494],"tight":[495],"CIs):":[496],"accuracy-positive":[501],"8/8":[503],"label-efficient":[508],"8/8,":[510],"honestly-reported":[513],"exception":[514],"near-chance":[516],"sleep/respiratory":[517],"task":[518],"cannot":[522],"amplify":[523],"absent":[524],"signal).":[525],"clean-accuracy":[527],"check":[528],"shows":[530,1083],"expensive":[532],"tax":[538,579,1062],"modestly":[540],"corruption-specifically,":[542],"while":[543,608,746,1101,1263],"ladder's":[545,1071,1180],"climbs":[547],"partition":[549],"depth":[550],"(one":[551],"channel":[554,1113],"rivals/edges":[555],"panel,":[558],"bit":[562,978],"cost)":[563],"dually":[566],"single":[569,761,805,1117,1167,1281],"drives":[571],"K":[572],"parallel":[573],"decisions":[574],"negligible":[577],"multi-task":[578],"(mean":[580],"+0.009":[581],"AUROC)":[582],"one-Kth":[584],"emitted":[586,727],"rate":[587],"sub-milliwatt":[589,808,1203],"core":[590,809],"multiplexing":[591],"several":[592],"detectors":[593],"price":[596],"one);":[598],"panel's":[606],"beating":[609],"noise":[612,791,1229,1317],"fraction":[616],"cost.":[619,981],"do":[622,1129,1194],"simply":[624],"stack:":[625],"composition":[626,1177,1373],"threat-dependent,":[628],"tempting":[630],"combination":[631],"(ladder":[632],"+":[633,1169],"spiral)":[634],"actively":[635],"conflicts":[636],"compound":[638,1163],"stress,":[639],"reproduced":[644,1145],"second":[647,689],"domain":[650,1052],"(PTB-XL":[651],"12-lead)":[652],"best":[655,1134],"configuration":[656],"threat-specific":[658],"running":[661],"all":[662],"together":[664],"worst":[667],"point":[668],"frontier":[671,818],"spiral-dominated":[673],"sum":[674],"costs,":[676],"1.4–3.3":[677],"ms,":[678],"most":[681],"+0.025":[683],"bump":[684],"plateaus":[686],"after":[687],"co-channel),":[690],"rule":[692],"being":[693],"compose":[695],"orthogonal":[697],"axes,":[698],"stack":[700,1131],"within":[701],"one.":[702],"robust":[706],"corruptions":[709,1201],"low-power":[711],"imposes":[714],"(powerline":[715],"interference,":[716],"int8":[717,1218],"quantization)":[718],"zero":[720,1221],"added":[721,1222],"cost;":[722],"identity":[723,745,1244,1262],"leakage":[724,1245],"falls":[729,1246],"toward":[730,1247],"chance":[731,1248],"grows":[735,1252],"entropy":[737,1254],"cap),":[738,1255],"adds":[743,1260],"no":[744,837,1225,1261],"identity-rich":[748,1265],"does;":[751],"honestly":[754],"when":[756],"unconstrained":[759,1279],"fixed":[762,1282],"dial":[763,778,1190,1226,1283,1303],"can":[764,1284],"equal":[765,1285],"best-per-threat":[767,1287],"oracle,":[768],"map's":[771,1273],"value":[772,802],"precisely":[774,1295],"cost-constrained:":[775],"meets":[780],"requirement.":[782,1307],"Findings":[783,1308],"are":[784,1236,1309],"stable":[785,1310],"readout,":[788],"codebook-size,":[789],"settings.":[792,1319],"recipe":[795],"follows":[797],"platform's":[800],"actual":[801],"proposition:":[803],"auditable,":[807],"partner":[812],"positions":[813],"anywhere":[814],"cost↔capability":[817,944,1339],"site,":[821],"threat,":[823],"power":[825],"budget.":[826],"characterization":[830],"previously-described,":[832],"filed":[833],"methods;":[834],"discloses":[836],"new":[838],"algorithmic":[839],"subject":[840],"matter,":[841],"per-deployment":[844],"selection":[845],"their":[849],"settings":[850],"retained":[852],"trade":[854],"secret.":[855],"Highlights":[856],"reframe:":[860],"story":[867],"story.":[873],"Answering":[874],"bluntest":[876],"reviewer":[877],"attack":[878],"(\\"did":[879],"you":[880],"just":[881],"add":[882],"random":[884],"forest?\\")":[885],"measurement:":[887],"panel":[891],"token+co-channel":[897],"moat":[906],"µs/window":[912],"(panel)":[915],"(full":[918],"model),":[919],"10–60×":[920],"keeps":[931],"off":[935],"One":[939],"core,":[940],"dials,":[942],"frontier.":[945],"previously-filed/-published":[961],"accuracy,":[964],"drift-":[965],"noise-robustness,":[967],"fault-tolerance,":[968],"label-efficiency":[970,1032],"/":[977,979,1230,1233,1432],"requires.":[989],"trade-off":[993],"pattern.":[999],"Across":[1000],"domains:":[1002],"workhorse":[1012],"(+0.05–0.19":[1013],"~16":[1015],"µs,":[1016],"~65%":[1017],"spiral's":[1020],"~1/60th":[1023],"compute);":[1025],"(D6)":[1029],"(population":[1033],"ceiling":[1034],"fresh":[1040],"chance,":[1044],"+0.33–0.48);":[1045],"(D1)":[1048],"helps":[1049],"every":[1050],"non-saturated":[1051],"free;":[1054],"(D2)":[1057],"(+0.12);":[1063],"(D5)":[1066],"additive-noise-specific.":[1068],"fault-tolerance.":[1075],"On":[1076],"saturated":[1078],"single-condition":[1079],"bench":[1080],"nothing":[1084],"collapses":[1093],"(macro-AUC":[1094],"0.70":[1095],"CWRU,":[1097],"0.56":[1098],"PTB-XL)":[1100],"channel-ladder":[1103],"(1.00":[1105],"+0.062):":[1107],"distributed":[1109],"multi-token":[1110],"code":[1111],"survives":[1112],"loss":[1114],"cannot.":[1119],"Reproduced":[1120],"both":[1122],"domains.":[1125],"choice":[1135],"threat-specific.":[1137],"2×2":[1140],"drift":[1142,1155],"×":[1143],"sensor-dropout,":[1144],"CWRU":[1147],"PTB-XL":[1149],"12-lead:":[1150],"clean":[1151],"token;":[1154],"free":[1157],"soft-vote;":[1158],"fault-tolerant":[1161],"ladder;":[1162],"drift+dropout":[1164],"which":[1173],"beats":[1174],"ladder+spiral":[1175],"conflict:":[1178],"drift-hit":[1181],"surviving-channel":[1182],"corrupts":[1184],"equal-weight":[1186],"fusion).":[1187],"Select":[1188],"threat;":[1193],"stack.":[1196],"Robust":[1198],"sees,":[1206],"in-frame.":[1209],"Inherent":[1210],"powerline":[1213],"interference":[1214],"(60":[1215],"Hz)":[1216],"quantization":[1219],"needed;":[1227],"broadband":[1228],"baseline":[1231],"wander":[1232],"missing":[1234],"samples":[1235],"rescued":[1237],"soft-vote.":[1242],"Emitted-token":[1243],"does.":[1268],"Honest":[1270],"about":[1271],"own":[1274],"value.":[1275],"When":[1276],"oracle":[1288],"map":[1292],"valuable":[1294],"budget:":[1298],"names":[1300],"clears":[1305],"(m),":[1313],"codebook-size":[1314],"(K),":[1315],"(σ)":[1318],"this":[1321],"record":[1322],"contains":[1323],"Manuscript_Paper47.pdf":[1325],"manuscript":[1328],"figures":[1331],"embedded,":[1332],"reading":[1334],"order":[1335],"(Figure":[1336],"1":[1337],"frontier,":[1340],"Figure":[1341,1346,1351,1357,1363,1369,1376],"2":[1342],"measured-cost":[1344],"bars,":[1345],"3":[1347],"data-movement/energy-savings":[1349],"distribution,":[1350],"4":[1352],"cross-domain":[1354],"heatmap,":[1356],"5":[1358],"cross-modality":[1360],"chart,":[1362],"6":[1364],"partition-depth":[1367],"curve,":[1368],"7":[1370],"two-domain":[1372],"map,":[1374],"8":[1377],"consolidated":[1379],"D.0–D7":[1380],"scorecard),":[1381],"Manuscript_Paper47.docx,":[1383],"editable":[1385],"source.":[1386],"PAPER_47_ZENODO_ARCHIVE.zip":[1388],"reproducibility":[1391],"archive":[1392],"(md5":[1393],"ARCHIVE_MD5.txt):":[1395],"eleven":[1397],"frozen":[1398],"pre-registrations":[1399,1414],"dial-map":[1401,1431],"R0–R7":[1404],"reviewer-proofing":[1405],"battery,":[1406],"Tier-2":[1408],"pre-registrations,":[1410],"refinement":[1413],"clean-accuracy,":[1417],"partition-depth/sketch,":[1419],"multi-task/parallel-decisions,":[1420],"scorecard-completion,":[1421],"dial-combinations,":[1422],"models),":[1429],"reviewer-pr":[1433]}

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-09-21
DOI
https://doi.org/10.5281/zenodo.22884022
Primary Topic
ECG Monitoring and Analysis
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
preprint

The Configurable Bottleneck: A Cost-Constrained Capability Map for a Deterministic Edge Decision Token

Randolph James Ferlic, Kimberly Kate Ferlic
Zenodo (CERN European Organization for Nuclear Research)
ECG Monitoring and Analysis
preprint

The Configurable Bottleneck: A Cost-Constrained Capability Map for a Deterministic Edge Decision Token

Randolph James Ferlic, Kimberly Kate Ferlic
preprint en

Abstract

The Configurable Bottleneck: A Cost-Constrained Capability Map for a Deterministic Edge Decision Token Randolph James Ferlic, M.D. and Kimberly Kate Ferlic (Fieldstone Analytics, LLC, Austin, TX, USA) Preprint · Zenodo DOI: 10.5281/zenodo.22884023 · CC-BY 4.0 · Community: spiral-domain-encoder-campaign Abstract A frozen, deterministic, class-discriminant single-token encoder (features → Fisher-discriminant ⊕ PCA subspace → k-means codebook of ≤ 256 cells → nearest-centroid decision) has been characterized piecemeal across a program of papers — its deployment behavior, its accuracy tax, its label-efficiency, its privacy — but never as one configurable platform, on one protocol, on the axes a deployment engineer actually trades. This is that unified, pre-registered, multi-axis, multi-domain characterization: a base token (M0) plus seven dials — a soft-vote readout (D1), a channel-partition token ladder (D2), a spiral (D3) and a diverse-model (D4) co-channel, a noise-robust reader (D5), a foundation codebook with shrinkage (D6), and a cheap quantized-sketch co-channel (D7), each a previously-filed or -published method — measured against the same base across ECG, industrial vibration, surgical-robot kinematics, wearable EMG, financial volatility, and continuous glucose data. The headline is deliberately honest and it reframes the object: on accuracy the token does not win — a cheap random-forest ensemble on the same features matches token-plus-co-channel, and a gradient-boosted model wins outright. What the token wins, by direct measurement, is cost: it reaches a decision in 3–5 microseconds per window versus 40–190 for that ensemble and 20–300 for the full model — 10 to 60× cheaper — as a fully enumerable ≤ 256-entry table that is auditable and never puts the raw signal on the wire. That silence on the wire is the token's largest cost advantage: against a full-data-stream-retention architecture it emits one byte instead of the raw window (an exact 180×–24,000× data reduction), and a pre-registered Monte-Carlo of radio-dominated device energy puts the transmission-energy savings at a median ≈ 350× (a Markov duty-cycle model brackets it from ≈ 14× vs an event-triggered device to ≈ 870× vs a continuous-streaming one; ≈ 900× even against the strongest decision-preserving codec) — one to two orders of magnitude beyond the compute figure. The platform is therefore a cost-versus-capability frontier: the bare token is the cheapest decision floor and concedes a real accuracy tax; the seven dials buy accuracy and robustness back, each at a stated compute, bit, or privacy cost, so a deployment spends only what its accuracy and threat budget require. Each dial's benefit generalizes across the six applications with a clear per-dial pattern (the diverse-model co-channel is the universal cheap accuracy/noise dial, +0.05–0.19; the foundation codebook is universal label-efficiency, near-ceiling at eight labels where a from-scratch codebook is at chance; the ladder recovers the multichannel accuracy tax, +0.12, and — its real headline — holds under sensor dropout where the bare token collapses; the noise-reader is additive-noise-specific). This generalization is confirmed by a pre-registered breadth battery on seven further physically-distinct modalities — inertial, chemical, acoustic/speech, gait, surgical-robotics, sleep/respiratory, and EEG — and at population scale (21,699 ECG records, tight CIs): the diverse-model co-channel is accuracy-positive on 8/8 and the foundation codebook label-efficient on 8/8, with one honestly-reported exception (a near-chance sleep/respiratory task where the dials cannot amplify absent signal). A clean-accuracy check further shows the expensive spiral co-channel recovers the accuracy tax only modestly and corruption-specifically, while the ladder's accuracy climbs with partition depth (one token per channel rivals/edges the diverse-model panel, at a stated bit cost) and — dually — a single token drives K parallel decisions at a negligible multi-task tax (mean +0.009 AUROC) and one-Kth the emitted rate (a sub-milliwatt core multiplexing several detectors for the price of one); and a cheap quantized-sketch co-channel matches that panel's accuracy while beating it on noise robustness at a fraction of the cost. The dials do not simply stack: composition is threat-dependent, one tempting combination (ladder + spiral) actively conflicts under compound stress, and — pre-registered and reproduced on a second raw multichannel domain (PTB-XL 12-lead) — the best configuration is threat-specific — and running all dials together is the worst point on the frontier (a spiral-dominated sum of costs, 1.4–3.3 ms, for at most a +0.025 bump that plateaus after the second co-channel), the rule being to compose across orthogonal axes, never stack within one. The token is robust to the corruptions a low-power sensor actually imposes (powerline interference, int8 quantization) at zero added cost; identity leakage from the emitted token falls toward chance as the population grows (an entropy cap), and a decision co-channel adds no identity while an identity-rich raw one does; and — honestly — when compute is unconstrained a single fixed dial can equal the best-per-threat oracle, so the map's value is precisely cost-constrained: the cheapest dial that meets each requirement. Findings are stable across the readout, codebook-size, and noise settings. The deployment recipe that follows is the platform's actual value proposition: a single deterministic, auditable, sub-milliwatt core that a partner positions anywhere on the cost↔capability frontier — per site, per threat, per power budget. This is a characterization of previously-described, filed methods; it discloses no new algorithmic subject matter, and the per-deployment selection of dials and their settings is retained as trade secret. Highlights · The honest reframe: the token is not the accuracy story — it is the cost story. Answering the bluntest reviewer attack ("did you just add a random forest?") with measurement: a cheap diverse-model panel on the same features matches token+co-channel and a gradient-boosted model wins outright. The token's moat is measured cost — 3–5 µs/window versus 40–190 (panel) and 20–300 (full model), 10–60× cheaper — as an enumerable ≤ 256-entry auditable table that keeps the raw signal off the wire. · One core, seven dials, one cost↔capability frontier. The bare token is the cheapest decision floor and concedes a real accuracy tax; seven previously-filed/-published dials buy accuracy, drift- and noise-robustness, fault-tolerance, and label-efficiency back, each at a stated compute / bit / privacy cost. A deployment spends only what its budget requires. · Each dial's trade-off generalizes — with a per-dial pattern. Across six domains: the diverse-model co-channel (D4) is the universal cheap accuracy/noise workhorse (+0.05–0.19 at ~16 µs, ~65% of the spiral's benefit at ~1/60th its compute); the foundation codebook (D6) is universal label-efficiency (population ceiling at eight labels where a fresh codebook is at chance, +0.33–0.48); the soft-vote (D1) helps every non-saturated domain for free; the ladder (D2) recovers the multichannel accuracy tax (+0.12); the noise-reader (D5) is additive-noise-specific. · The ladder's real headline is fault-tolerance. On a saturated single-condition bench the ladder shows nothing — but under sensor dropout the bare token collapses (macro-AUC 0.70 on CWRU, 0.56 on PTB-XL) while the channel-ladder holds (1.00 and +0.062): the distributed multi-token code survives channel loss where the single token cannot. Reproduced on both raw multichannel domains. · The dials do not stack — the best choice is threat-specific. A pre-registered 2×2 of drift × sensor-dropout, reproduced on CWRU and PTB-XL 12-lead: clean → bare token; drift → free soft-vote; dropout → fault-tolerant ladder; compound drift+dropout → a single token + spiral co-channel — which beats ladder+spiral (the composition conflict: the ladder's drift-hit surviving-channel token corrupts the equal-weight fusion). Select one dial by the threat; do not stack. · Robust to the corruptions a sub-milliwatt sensor actually sees, and privacy in-frame. Inherent robustness to powerline interference (60 Hz) and int8 quantization at zero added cost and no dial needed; broadband noise / baseline wander / missing samples are rescued by a co-channel or soft-vote. Emitted-token identity leakage falls toward chance as the population grows (an entropy cap), and a decision co-channel adds no identity while an identity-rich raw one does. · Honest about the map's own value. When compute is unconstrained a single fixed dial can equal the best-per-threat oracle — so the map is valuable precisely under a budget: it names the cheapest dial that clears each requirement. Findings are stable across readout (m), codebook-size (K), and noise (σ) settings. What this record contains · Manuscript_Paper47.pdf — the manuscript with eight figures embedded, in reading order (Figure 1 the cost↔capability frontier, Figure 2 the measured-cost bars, Figure 3 the data-movement/energy-savings distribution, Figure 4 the cross-domain generalization heatmap, Figure 5 the cross-modality breadth chart, Figure 6 the ladder partition-depth curve, Figure 7 the two-domain composition map, and Figure 8 the consolidated D.0–D7 scorecard), and Manuscript_Paper47.docx, the editable source. · PAPER_47_ZENODO_ARCHIVE.zip — the reproducibility archive (md5 in ARCHIVE_MD5.txt): the eleven frozen pre-registrations (the dial-map protocol, the R0–R7 reviewer-proofing battery, two Tier-2 breadth pre-registrations, and seven refinement pre-registrations — spiral clean-accuracy, ladder partition-depth/sketch, multi-task/parallel-decisions, scorecard-completion, dial-combinations, and the Monte-Carlo and Markov energy models), the dial-map / reviewer-pr

Zenodo (CERN European Organization for Nuclear Research)
EP Analytics (United States) (US)
Peace, Justice and strong institutions, Reduced inequalities
ECG Monitoring and Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.