1249:: the choice of examples to label is seen as a dilemma between the exploration and the exploitation over the data space representation. This strategy manages this compromise by modelling the active learning problem as a contextual bandit problem. For example, Bouneffouf et al. propose a sequential algorithm named Active Thompson Sampling (ATS), which, in each round, assigns a sampling distribution on the pool, samples one point from this distribution, and queries the oracle for this sample point label.
983:
of examples to learn a concept can often be much lower than the number required in normal supervised learning. With this approach, there is a risk that the algorithm is overwhelmed by uninformative examples. Recent developments are dedicated to multi-label active learning, hybrid active learning and active learning in a single-pass (on-line) context, combining concepts from the field of machine learning (e.g. conflict and ignorance) with adaptive,
1351:
algorithms have been gaining in popularity. Some of them have been proposed to tackle the problem of learning AL strategies instead of relying on manually designed strategies. A benchmark which compares 'meta-learning approaches to active learning' to 'traditional heuristic-based Active
Learning' may give intuitions if 'Learning active learning' is at the crossroads
1212:: This is where the learner generates synthetic data from an underlying natural distribution. For example, if the dataset are pictures of humans and animals, the learner could send a clipped image of a leg to the teacher and query if this appendage belongs to an animal or human. This is particularly useful if the dataset is small.
1339:: The primary selection criterion is the prediction mismatch between the current model and nearest-neighbour prediction. It targets on wrongly predicted data points. The second selection criterion is the distance to previously selected data, the farthest first. It aims at optimizing the diversity of selected data.
1233:(AST) measure liver function (though AST is also produced by other tissues, e.g., lung, pancreas) A synthetic data point with AST at the lower limit of normal range (8-33 Units/L) with an ALT several times above normal range (4-35 Units/L) in a simulated chronically ill patient would be physiologically impossible.
1206:
algorithm does not have sufficient information, early in the process, to make a sound assign-label-vs ask-teacher decision, and it does not capitalize as efficiently on the presence of already labeled data. Therefore, the teacher is likely to spend more effort in supplying labels than with the pool-based approach.
1191:
before selecting data points (instances) for labeling. It is often initially trained on a fully labeled subset of the data using a machine-learning method such as logistic regression or SVM that yields class-membership probabilities for individual data instances. The candidate instances are those for
982:
There are situations in which unlabeled data is abundant but manual labeling is expensive. In such a scenario, learning algorithms can actively query the user/teacher for labels. This type of iterative supervised learning is called active learning. Since the learner chooses the examples, the number
1194:
The theoretical drawback of pool-based sampling is that it is memory-intensive and is therefore limited in its capacity to handle enormous datasets, but in practice, the rate-limiting factor is that the teacher is typically a (fatiguable) human expert who must be paid for their effort, rather than
1350:
A wide variety of algorithms have been studied that fall into these categories. While the traditional AL strategies can achieve remarkable performance, it is often challenging to predict in advance which strategy is the most suitable in aparticular situation. In recent years, meta-learning
1205:
with the machine evaluating the informativeness of each item against its query parameters. The learner decides for itself whether to assign a label or query the teacher for each datapoint. As contrasted with Pool-based sampling, the obvious drawback of stream-based methods is that the learning
1192:
which the prediction is most ambiguous. Instances are drawn from the entire data pool and assigned a confidence score, a measurement of how well the learner "understands" the data. The system then selects the instances for which it is the least confident and queries the teacher for the labels.
1214:
The challenge here, as with all synthetic-data-generation efforts, is in ensuring that the synthetic data is consistent in terms of meeting the constraints on real data. As the number of variables/features in the input data increase, and strong dependencies between variables exist, it becomes
966:
in which a learning algorithm can interactively query a human user (or some other information source), to label new data points with the desired outputs. The human user must possess knowledge/expertise in the problem domain, including the ability to consult/research authoritative sources when
1979:
Faria, Bruno; Perdigão, Dylan; Brás, Joana; Macedo, Luis (2022). "The Joint Role of Batch Size and Query
Strategy in Active Learning-Based Prediction - A Case Study in the Heart Attack Domain". In Goreti Marreiros; Bruno Martins; Ana Paiva; Bernardete Ribeiro; Alberto Sardinha (eds.).
1571:
Das, Shubhomoy; Wong, Weng-Keen; Dietterich, Thomas; Fern, Alan; Emmott, Andrew (2016). "Incorporating Expert
Feedback into Active Anomaly Discovery". In Bonchi, Francesco; Domingo-Ferrer, Josep; Baeza-Yates, Ricardo; Zhou, Zhi-Hua; Wu, Xindong (eds.).
2160:. Proceedings of the Workshop on Interactive Adaptive Learning co-located with European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases ({ECML} {PKDD} 2020), Ghent, Belgium, 2020.
1345:
Learning is accomplished by applying dimensionality reduction to graphs and figures like scatter plots. Then the user is asked to label the compiled data (categorical, numerical, relevance scores, relation between two
1333:: predicts that a new data point will have a label similar to old data points in some specified way and degree of the similarity within the old examples is used to estimate the confidence in the prediction.
1271:: In this paper, the author proposes a sequential algorithm named exponentiated gradient (EG)-active that can improve any active learning algorithm by an optimal random exploration.
1984:. 21st EPIA Conference on Artificial Intelligence, EPIA 2022, Lisbon, Portugal, August 31–September 2, 2022. Lecture Notes in Computer Science. Vol. 13566. pp. 464–475.
1151:
1110:
1069:
2037:
Makili, Lázaro EmĂlio; Sánchez, JesĂşs A. Vega; Dormido-Canto, Sebastián (2012-10-01). "Active
Learning Using Conformal Predictors: Application to Image Classification".
1517:
Rubens, Neil; Elahi, Mehdi; Sugiyama, Masashi; Kaplan, Dain (2016). "Active
Learning in Recommender Systems". In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.).
1309:: a variety of models are trained on the current labeled data, and vote on the output for unlabeled data; label those points for which the "committee" disagrees the most
873:
911:
2113:
Bernard, JĂĽrgen; Zeppelzauer, Matthias; Lehmann, Markus; MĂĽller, Martin; Sedlmair, Michael (June 2018). "Towards User-Centered Active
Learning Algorithms".
1022:
would include all proteins that are known to have a certain interesting activity and all additional proteins that one might want to test for that activity.
991:. Using active learning allows for faster development of a machine learning algorithm, when comparative updates would require a quantum or super computer.
868:
858:
699:
1242:
Algorithms for determining which data points should be labeled can be organized into a number of different categories, based upon their purpose:
1417:
906:
863:
714:
445:
946:
749:
1689:
Lughofer, Edwin (February 2012). "Hybrid active learning for reducing the annotation effort of operators in classification systems".
1450:
Balcan, Maria-Florina & Hanneke, Steve & Wortman, Jennifer. (2008). The True Sample
Complexity of Active Learning.. 45-56.
825:
1444:
Improving
Generalization with Active Learning, David Cohn, Les Atlas & Richard Ladner, Machine Learning 15, 201–221 (1994).
1876:
374:
1997:
1897:
1665:
1589:
1547:
1319:. This offers the possibility of selecting instances from non-overlapping or minimally overlapping partitions for labeling.
1295:
The sample with the smallest difference between the two highest class probabilities is considered to be the most uncertain.
883:
646:
181:
1315:: When the underlying model is a forest of trees, the leaf nodes might represent (overlapping) partitions of the original
1289:
The entropy formula is used on each sample, and the sample with the highest entropy is considered to be the least certain.
901:
2013:
1632:
1363:(SVMs) and exploit the structure of the SVM to determine which data points to label. Such methods usually calculate the
709:
658:
782:
777:
430:
440:
78:
1518:
939:
835:
599:
420:
1222:
810:
512:
288:
1640:
Proceedings of the 15th ACM SIGKDD international conference on
Knowledge discovery and data mining - KDD '09
1283:: label those points for which the current model is least certain as to what the correct output should be.
767:
704:
614:
592:
435:
425:
968:
918:
830:
815:
276:
98:
1400:
to be labeled. Other similar methods, such as
Maximum Marginal Hyperplane, choose data with the largest
1121:
1080:
1039:
1466:
Learning how to Active Learn: A Deep Reinforcement Learning Approach, Meng Fang, Yuan Li, Trevor Cohn,
1316:
1187:: In this approach, which is the most well known scenario, the learning algorithm attempts to evaluate
1168:
Most of the current research in active learning involves the best method to choose the data points for
878:
805:
555:
450:
238:
171:
131:
19:
This article is about a machine learning method. For active learning in the context of education, see
2185:
2088:
Zhao, Shuyang; Heittola, Toni; Virtanen, Tuomas (2020). "Active learning for sound event detection".
1364:
1217:
For example, to create a synthetic data set for human laboratory-test values, the sum of the various
932:
538:
306:
176:
1648:
1493:
1460:: a Unified Perspective to Learn with a Goal, Francesco Di Fiore, Michela Nardelli, Laura Mainini,
560:
480:
403:
321:
151:
113:
108:
68:
63:
1803:
1325:: label those points that would minimize output variance, which is one of the components of error.
1830:
1018:
be the total set of all data under consideration. For example, in a protein engineering problem,
988:
507:
356:
256:
83:
1871:
Bouneffouf, Djallel; Laroche, Romain; Urvoy, Tanguy; FĂ©raud, Raphael; Allesiardo, Robin (2014).
1643:
1432:
1360:
1230:
999:
687:
663:
565:
326:
301:
261:
73:
1618:"A literature survey of active machine learning in the context of natural language processing"
1451:
1457:
1427:
641:
463:
415:
271:
186:
58:
2046:
1698:
1329:
1262:
1226:
1225:
must equal 100, since the component numbers are really percentages. Similarly, the enzymes
984:
570:
520:
8:
673:
609:
580:
485:
311:
244:
230:
216:
191:
141:
93:
53:
2050:
1702:
2161:
2138:
2093:
2070:
1954:
1934:
1903:
1853:
1772:
1741:
1671:
1595:
1553:
651:
575:
361:
156:
2165:
2130:
2074:
2062:
1993:
1893:
1661:
1585:
1543:
1422:
1301:
The sample with the smallest best probability is considered to be the most uncertain.
1003:
744:
587:
500:
296:
266:
211:
206:
161:
103:
2142:
1958:
1745:
1599:
1557:
2122:
2054:
1985:
1944:
1907:
1885:
1857:
1845:
1782:
1733:
1724:
Lughofer, Edwin (2012). "Single-pass active learning with conflict and ignorance".
1706:
1675:
1653:
1577:
1535:
1527:
1218:
963:
772:
525:
475:
385:
369:
339:
201:
196:
146:
136:
34:
1989:
1889:
1849:
1710:
1393:
are those that the SVM is most uncertain about and therefore should be placed in
800:
604:
470:
410:
20:
1872:
1787:
1760:
820:
351:
88:
1737:
1531:
2179:
2134:
2066:
1949:
1922:
995:
739:
668:
550:
281:
166:
1657:
1215:
increasingly difficult to generate synthetic data with sufficient fidelity.
1581:
1389:
Minimum Marginal Hyperplane methods assume that the data with the smallest
2058:
1499:. Computer Sciences Technical Report 1648. University of Wisconsin–Madison
1761:"The MLIP package: moment tensor potentials with MPI and active learning"
1539:
545:
39:
1884:. Lecture Notes in Computer Science. Vol. 8834. pp. 405–412.
694:
390:
316:
2126:
1875:. In Loo, C. K.; Yap, K. S.; Wong, K. W.; Teoh, A.; Huang, K. (eds.).
2158:
Learning Active Learning at the Crossroads? Evaluation and Discussion
1467:
1461:
1445:
853:
634:
1386:-dimensional distance from that datum to the separating hyperplane.
2098:
1777:
1939:
1831:"Active learning via query synthesis and nearest neighbour search"
967:
necessary. In statistics literature, it is sometimes also called
1631:
Yang, Bishan; Sun, Jian-Tao; Wang, Tengjiao; Chen, Zheng (2009).
629:
1829:
Wang, Liantao; Hu, Xuelei; Yuan, Bo; Lu, Jianfeng (2015-01-05).
2112:
2090:
IEEE/ACM Transactions on Audio, Speech, and Language Processing
1804:"Active learning machine learning: What it is and how it works"
1633:"Effective multi-label active learning for text classification"
1617:
380:
1255:: label those points that would most change the current model.
624:
619:
346:
1870:
1404:. Tradeoff methods choose a mix of the smallest and largest
1452:
https://link.springer.com/article/10.1007/s10994-010-5174-y
2036:
1516:
1923:"Exponentiated Gradient Exploration for Active Learning"
1873:"Contextual Bandit for Active Learning: Active Thompson"
1261:: label those points that would most reduce the model's
1201:: Here, each consecutive unlabeled instance is examined
912:
List of datasets in computer vision and image processing
1978:
1269:
Exponentiated Gradient Exploration for Active Learning
994:
Large-scale active learning projects may benefit from
1570:
1124:
1083:
1042:
2087:
2006:
1145:
1104:
1063:
2155:
1574:IEEE 16th International Conference on Data Mining
2177:
1914:
1864:
1630:
1615:
1359:Some active learning algorithms are built upon
1418:List of datasets for machine learning research
1354:
907:List of datasets for machine-learning research
1974:
1972:
1970:
1968:
1491:
1313:Querying from diverse subspaces or partitions
940:
2156:Desreumaux, Louis; Lemaire, Vincent (2020).
1828:
1965:
1920:
1717:
947:
933:
2097:
1948:
1938:
1786:
1776:
1647:
1510:
1487:
1485:
1483:
1765:Machine Learning: Science and Technology
1723:
1688:
1611:
1609:
971:. The information source is also called
1758:
2178:
1921:Bouneffouf, Djallel (8 January 2016).
1564:
1480:
2081:
1801:
1682:
1606:
1624:
1247:Balance exploration and exploitation
1982:Progress in Artificial Intelligence
1494:"Active Learning Literature Survey"
1237:
902:Glossary of artificial intelligence
13:
2106:
1468:https://arxiv.org/abs/1708.02383v1
1462:https://arxiv.org/abs/2303.01560v2
1446:https://doi.org/10.1007/BF00993277
1343:User Centered Labeling Strategies:
1146:{\displaystyle \mathbf {T} _{C,i}}
1105:{\displaystyle \mathbf {T} _{U,i}}
1064:{\displaystyle \mathbf {T} _{K,i}}
1004:humans in the active learning loop
14:
2197:
1620:. SICS Technical Report T2009:06.
1337:Mismatch-first farthest-traversal
1112:: Data points where the label is
1071:: Data points where the label is
1127:
1086:
1045:
1033:is broken up into three subsets
2149:
2030:
1822:
1199:Stream-Based Selective Sampling
1795:
1752:
1616:Olsson, Fredrik (April 2009).
1277:a sample is randomly selected.
1009:
322:Relevance vector machine (RVM)
1:
2039:Fusion Science and Technology
1878:Neural Information Processing
1473:
1438:
1371:, of each unlabeled datum in
1223:White Blood Cell differential
811:Computational learning theory
375:Expectation–maximization (EM)
1990:10.1007/978-3-031-16474-3_38
1890:10.1007/978-3-319-12637-1_51
1850:10.1016/j.neucom.2014.06.042
1711:10.1016/j.patcog.2011.08.009
1520:Recommender Systems Handbook
1178:
768:Coefficient of determination
615:Convolutional neural network
327:Support vector machine (SVM)
7:
1526:(2 ed.). Springer US.
1411:
1355:Minimum marginal hyperplane
969:optimal experimental design
919:Outline of machine learning
816:Empirical risk minimization
10:
2202:
2014:"shubhomoydas/ad_examples"
1576:. IEEE. pp. 853–858.
1210:Membership Query Synthesis
556:Feedforward neural network
307:Artificial neural networks
18:
1738:10.1007/s12530-012-9060-7
1532:10.1007/978-1-4899-7637-6
1299:Least Confident Sampling:
987:policies in the field of
539:Artificial neural network
16:Machine learning strategy
1950:10.3390/computers5010001
1788:10.1088/2632-2153/abc9fe
1259:Expected error reduction
848:Journals and conferences
795:Mathematical foundations
705:Temporal difference (TD)
561:Recurrent neural network
481:Conditional random field
404:Dimensionality reduction
152:Dimensionality reduction
114:Quantum machine learning
109:Neuromorphic engineering
69:Self-supervised learning
64:Semi-supervised learning
2115:Computer Graphics Forum
1910:. HAL Id: hal-01069802.
1658:10.1145/1557019.1557119
1361:support-vector machines
1025:During each iteration,
989:online machine learning
257:Apprenticeship learning
1759:Novikov, Ivan (2021).
1582:10.1109/ICDM.2016.0102
1492:Settles, Burr (2010).
1433:Reinforcement learning
1231:Aspartate Transaminase
1221:(WBC) components in a
1147:
1106:
1065:
1000:Amazon Mechanical Turk
806:Bias–variance tradeoff
688:Reinforcement learning
664:Spiking neural network
74:Reinforcement learning
2059:10.13182/FST12-A14626
1458:Bayesian Optimization
1428:Bayesian Optimization
1253:Expected model change
1148:
1107:
1066:
962:is a special case of
642:Neural radiance field
464:Structured prediction
187:Structured prediction
59:Unsupervised learning
1456:Active Learning and
1330:Conformal prediction
1281:Uncertainty sampling
1263:generalization error
1227:Alanine Transaminase
1122:
1081:
1040:
985:incremental learning
831:Statistical learning
729:Learning with humans
521:Local outlier factor
2051:2012FuST...62..347M
1703:2012PatRe..45..884L
1691:Pattern Recognition
1185:Pool-Based Sampling
998:frameworks such as
674:Electrochemical RAM
581:reservoir computing
312:Logistic regression
231:Supervised learning
217:Multimodal learning
192:Feature engineering
137:Generative modeling
99:Rule-based learning
94:Curriculum learning
54:Supervised learning
29:Part of a series on
1323:Variance reduction
1307:Query by committee
1189:the entire dataset
1143:
1102:
1061:
1002:that include many
242: •
157:Density estimation
2127:10.1111/cgf.13406
1999:978-3-031-16473-6
1899:978-3-319-12636-4
1667:978-1-60558-495-9
1591:978-1-5090-5473-2
1549:978-1-4899-7637-6
1423:Sample complexity
1287:Entropy Sampling:
957:
956:
762:Model diagnostics
745:Human-in-the-loop
588:Boltzmann machine
501:Anomaly detection
297:Linear regression
212:Ontology learning
207:Grammar induction
182:Semantic analysis
177:Association rules
162:Anomaly detection
104:Neuro-symbolic AI
2193:
2186:Machine learning
2170:
2169:
2153:
2147:
2146:
2110:
2104:
2103:
2101:
2085:
2079:
2078:
2034:
2028:
2027:
2025:
2024:
2010:
2004:
2003:
1976:
1963:
1962:
1952:
1942:
1918:
1912:
1911:
1883:
1868:
1862:
1861:
1835:
1826:
1820:
1819:
1817:
1815:
1799:
1793:
1792:
1790:
1780:
1756:
1750:
1749:
1726:Evolving Systems
1721:
1715:
1714:
1686:
1680:
1679:
1651:
1637:
1628:
1622:
1621:
1613:
1604:
1603:
1568:
1562:
1561:
1525:
1514:
1508:
1507:
1505:
1504:
1498:
1489:
1407:
1403:
1399:
1392:
1385:
1381:
1377:
1370:
1293:Margin Sampling:
1275:Random Sampling:
1238:Query strategies
1219:white blood cell
1195:computer memory.
1174:
1159:
1152:
1150:
1149:
1144:
1142:
1141:
1130:
1111:
1109:
1108:
1103:
1101:
1100:
1089:
1070:
1068:
1067:
1062:
1060:
1059:
1048:
1032:
1028:
1021:
1017:
964:machine learning
949:
942:
935:
896:Related articles
773:Confusion matrix
526:Isolation forest
471:Graphical models
250:
249:
202:Learning to rank
197:Feature learning
35:Machine learning
26:
25:
2201:
2200:
2196:
2195:
2194:
2192:
2191:
2190:
2176:
2175:
2174:
2173:
2154:
2150:
2111:
2107:
2086:
2082:
2035:
2031:
2022:
2020:
2012:
2011:
2007:
2000:
1977:
1966:
1919:
1915:
1900:
1881:
1869:
1865:
1833:
1827:
1823:
1813:
1811:
1810:. DataRobot Inc
1800:
1796:
1757:
1753:
1722:
1718:
1687:
1683:
1668:
1649:10.1.1.546.9358
1642:. p. 917.
1635:
1629:
1625:
1614:
1607:
1592:
1569:
1565:
1550:
1523:
1515:
1511:
1502:
1500:
1496:
1490:
1481:
1476:
1441:
1414:
1405:
1401:
1398:
1394:
1390:
1383:
1379:
1376:
1372:
1368:
1357:
1240:
1216:
1213:
1193:
1181:
1173:
1169:
1158:
1154:
1131:
1126:
1125:
1123:
1120:
1119:
1090:
1085:
1084:
1082:
1079:
1078:
1049:
1044:
1043:
1041:
1038:
1037:
1030:
1026:
1019:
1015:
1012:
960:Active learning
953:
924:
923:
897:
889:
888:
849:
841:
840:
801:Kernel machines
796:
788:
787:
763:
755:
754:
735:Active learning
730:
722:
721:
690:
680:
679:
605:Diffusion model
541:
531:
530:
503:
493:
492:
466:
456:
455:
411:Factor analysis
406:
396:
395:
379:
342:
332:
331:
252:
251:
235:
234:
233:
222:
221:
127:
119:
118:
84:Online learning
49:
37:
24:
21:active learning
17:
12:
11:
5:
2199:
2189:
2188:
2172:
2171:
2148:
2121:(3): 121–132.
2105:
2080:
2045:(2): 347–355.
2029:
2005:
1998:
1964:
1913:
1898:
1863:
1838:Neurocomputing
1821:
1808:DataRobot Blog
1794:
1751:
1732:(4): 251–271.
1716:
1697:(2): 884–896.
1681:
1666:
1623:
1605:
1590:
1563:
1548:
1509:
1478:
1477:
1475:
1472:
1471:
1470:
1464:
1454:
1448:
1440:
1437:
1436:
1435:
1430:
1425:
1420:
1413:
1410:
1396:
1374:
1356:
1353:
1348:
1347:
1340:
1334:
1326:
1320:
1310:
1304:
1303:
1302:
1296:
1290:
1278:
1272:
1266:
1256:
1250:
1239:
1236:
1235:
1234:
1207:
1196:
1180:
1177:
1171:
1166:
1165:
1164:to be labeled.
1156:
1153:: A subset of
1140:
1137:
1134:
1129:
1117:
1099:
1096:
1093:
1088:
1076:
1058:
1055:
1052:
1047:
1011:
1008:
955:
954:
952:
951:
944:
937:
929:
926:
925:
922:
921:
916:
915:
914:
904:
898:
895:
894:
891:
890:
887:
886:
881:
876:
871:
866:
861:
856:
850:
847:
846:
843:
842:
839:
838:
833:
828:
823:
821:Occam learning
818:
813:
808:
803:
797:
794:
793:
790:
789:
786:
785:
780:
778:Learning curve
775:
770:
764:
761:
760:
757:
756:
753:
752:
747:
742:
737:
731:
728:
727:
724:
723:
720:
719:
718:
717:
707:
702:
697:
691:
686:
685:
682:
681:
678:
677:
671:
666:
661:
656:
655:
654:
644:
639:
638:
637:
632:
627:
622:
612:
607:
602:
597:
596:
595:
585:
584:
583:
578:
573:
568:
558:
553:
548:
542:
537:
536:
533:
532:
529:
528:
523:
518:
510:
504:
499:
498:
495:
494:
491:
490:
489:
488:
483:
478:
467:
462:
461:
458:
457:
454:
453:
448:
443:
438:
433:
428:
423:
418:
413:
407:
402:
401:
398:
397:
394:
393:
388:
383:
377:
372:
367:
359:
354:
349:
343:
338:
337:
334:
333:
330:
329:
324:
319:
314:
309:
304:
299:
294:
286:
285:
284:
279:
274:
264:
262:Decision trees
259:
253:
239:classification
229:
228:
227:
224:
223:
220:
219:
214:
209:
204:
199:
194:
189:
184:
179:
174:
169:
164:
159:
154:
149:
144:
139:
134:
132:Classification
128:
125:
124:
121:
120:
117:
116:
111:
106:
101:
96:
91:
89:Batch learning
86:
81:
76:
71:
66:
61:
56:
50:
47:
46:
43:
42:
31:
30:
15:
9:
6:
4:
3:
2:
2198:
2187:
2184:
2183:
2181:
2167:
2163:
2159:
2152:
2144:
2140:
2136:
2132:
2128:
2124:
2120:
2116:
2109:
2100:
2095:
2091:
2084:
2076:
2072:
2068:
2064:
2060:
2056:
2052:
2048:
2044:
2040:
2033:
2019:
2015:
2009:
2001:
1995:
1991:
1987:
1983:
1975:
1973:
1971:
1969:
1960:
1956:
1951:
1946:
1941:
1936:
1932:
1928:
1924:
1917:
1909:
1905:
1901:
1895:
1891:
1887:
1880:
1879:
1874:
1867:
1859:
1855:
1851:
1847:
1843:
1839:
1832:
1825:
1809:
1805:
1798:
1789:
1784:
1779:
1774:
1770:
1766:
1762:
1755:
1747:
1743:
1739:
1735:
1731:
1727:
1720:
1712:
1708:
1704:
1700:
1696:
1692:
1685:
1677:
1673:
1669:
1663:
1659:
1655:
1650:
1645:
1641:
1634:
1627:
1619:
1612:
1610:
1601:
1597:
1593:
1587:
1583:
1579:
1575:
1567:
1559:
1555:
1551:
1545:
1541:
1540:11311/1006123
1537:
1533:
1529:
1522:
1521:
1513:
1495:
1488:
1486:
1484:
1479:
1469:
1465:
1463:
1459:
1455:
1453:
1449:
1447:
1443:
1442:
1434:
1431:
1429:
1426:
1424:
1421:
1419:
1416:
1415:
1409:
1387:
1366:
1362:
1352:
1344:
1341:
1338:
1335:
1332:
1331:
1327:
1324:
1321:
1318:
1317:feature space
1314:
1311:
1308:
1305:
1300:
1297:
1294:
1291:
1288:
1285:
1284:
1282:
1279:
1276:
1273:
1270:
1267:
1264:
1260:
1257:
1254:
1251:
1248:
1245:
1244:
1243:
1232:
1228:
1224:
1220:
1211:
1208:
1204:
1203:one at a time
1200:
1197:
1190:
1186:
1183:
1182:
1176:
1163:
1138:
1135:
1132:
1118:
1115:
1097:
1094:
1091:
1077:
1074:
1056:
1053:
1050:
1036:
1035:
1034:
1023:
1007:
1005:
1001:
997:
996:crowdsourcing
992:
990:
986:
980:
978:
974:
970:
965:
961:
950:
945:
943:
938:
936:
931:
930:
928:
927:
920:
917:
913:
910:
909:
908:
905:
903:
900:
899:
893:
892:
885:
882:
880:
877:
875:
872:
870:
867:
865:
862:
860:
857:
855:
852:
851:
845:
844:
837:
834:
832:
829:
827:
824:
822:
819:
817:
814:
812:
809:
807:
804:
802:
799:
798:
792:
791:
784:
781:
779:
776:
774:
771:
769:
766:
765:
759:
758:
751:
748:
746:
743:
741:
740:Crowdsourcing
738:
736:
733:
732:
726:
725:
716:
713:
712:
711:
708:
706:
703:
701:
698:
696:
693:
692:
689:
684:
683:
675:
672:
670:
669:Memtransistor
667:
665:
662:
660:
657:
653:
650:
649:
648:
645:
643:
640:
636:
633:
631:
628:
626:
623:
621:
618:
617:
616:
613:
611:
608:
606:
603:
601:
598:
594:
591:
590:
589:
586:
582:
579:
577:
574:
572:
569:
567:
564:
563:
562:
559:
557:
554:
552:
551:Deep learning
549:
547:
544:
543:
540:
535:
534:
527:
524:
522:
519:
517:
515:
511:
509:
506:
505:
502:
497:
496:
487:
486:Hidden Markov
484:
482:
479:
477:
474:
473:
472:
469:
468:
465:
460:
459:
452:
449:
447:
444:
442:
439:
437:
434:
432:
429:
427:
424:
422:
419:
417:
414:
412:
409:
408:
405:
400:
399:
392:
389:
387:
384:
382:
378:
376:
373:
371:
368:
366:
364:
360:
358:
355:
353:
350:
348:
345:
344:
341:
336:
335:
328:
325:
323:
320:
318:
315:
313:
310:
308:
305:
303:
300:
298:
295:
293:
291:
287:
283:
282:Random forest
280:
278:
275:
273:
270:
269:
268:
265:
263:
260:
258:
255:
254:
247:
246:
241:
240:
232:
226:
225:
218:
215:
213:
210:
208:
205:
203:
200:
198:
195:
193:
190:
188:
185:
183:
180:
178:
175:
173:
170:
168:
167:Data cleaning
165:
163:
160:
158:
155:
153:
150:
148:
145:
143:
140:
138:
135:
133:
130:
129:
123:
122:
115:
112:
110:
107:
105:
102:
100:
97:
95:
92:
90:
87:
85:
82:
80:
79:Meta-learning
77:
75:
72:
70:
67:
65:
62:
60:
57:
55:
52:
51:
45:
44:
41:
36:
33:
32:
28:
27:
22:
2157:
2151:
2118:
2114:
2108:
2089:
2083:
2042:
2038:
2032:
2021:. Retrieved
2017:
2008:
1981:
1930:
1926:
1916:
1877:
1866:
1841:
1837:
1824:
1812:. Retrieved
1807:
1797:
1768:
1764:
1754:
1729:
1725:
1719:
1694:
1690:
1684:
1639:
1626:
1573:
1566:
1519:
1512:
1501:. Retrieved
1388:
1358:
1349:
1342:
1336:
1328:
1322:
1312:
1306:
1298:
1292:
1286:
1280:
1274:
1268:
1258:
1252:
1246:
1241:
1209:
1202:
1198:
1188:
1184:
1167:
1161:
1113:
1072:
1024:
1013:
993:
981:
976:
972:
959:
958:
826:PAC learning
734:
513:
362:
357:Hierarchical
289:
243:
237:
1844:: 426–434.
1802:DataRobot.
1771:(2): 3, 4.
1010:Definitions
710:Multi-agent
647:Transformer
546:Autoencoder
302:Naive Bayes
40:data mining
2099:2002.05033
2023:2018-12-04
1814:30 January
1778:2007.08555
1503:2014-11-18
1474:References
1439:Literature
1378:and treat
1346:instances.
1229:(ALT) and
695:Q-learning
593:Restricted
391:Mean shift
340:Clustering
317:Perceptron
245:regression
147:Clustering
142:Regression
2166:221794570
2135:0167-7055
2075:115384000
2067:1536-1055
1940:1408.2196
1927:Computers
1644:CiteSeerX
1179:Scenarios
854:ECML PKDD
836:VC theory
783:ROC curve
715:Self-play
635:DeepDream
476:Bayes net
267:Ensembles
48:Paradigms
2180:Category
2143:51875861
1959:14313852
1933:(1): 1.
1746:43844282
1600:15285595
1558:11569603
1412:See also
1160:that is
277:Boosting
126:Problems
2047:Bibcode
1908:1701357
1858:3027214
1699:Bibcode
1676:1979173
1114:unknown
973:teacher
859:NeurIPS
676:(ECRAM)
630:AlexNet
272:Bagging
2164:
2141:
2133:
2073:
2065:
2018:GitHub
1996:
1957:
1906:
1896:
1856:
1744:
1674:
1664:
1646:
1598:
1588:
1556:
1546:
1382:as an
1365:margin
1162:chosen
977:oracle
652:Vision
508:RANSAC
386:OPTICS
381:DBSCAN
365:-means
172:AutoML
2162:S2CID
2139:S2CID
2094:arXiv
2071:S2CID
1955:S2CID
1935:arXiv
1904:S2CID
1882:(PDF)
1854:S2CID
1834:(PDF)
1773:arXiv
1742:S2CID
1672:S2CID
1636:(PDF)
1596:S2CID
1554:S2CID
1524:(PDF)
1497:(PDF)
1073:known
874:IJCAI
700:SARSA
659:Mamba
625:LeNet
620:U-Net
446:t-SNE
370:Fuzzy
347:BIRCH
2131:ISSN
2063:ISSN
1994:ISBN
1894:ISBN
1816:2024
1662:ISBN
1586:ISBN
1544:ISBN
1014:Let
884:JMLR
869:ICLR
864:ICML
750:RLHF
566:LSTM
352:CURE
38:and
2123:doi
2055:doi
1986:doi
1945:doi
1886:doi
1846:doi
1842:147
1783:doi
1734:doi
1707:doi
1654:doi
1578:doi
1536:hdl
1528:doi
1408:s.
1397:C,i
1375:U,i
1172:C,i
1157:U,i
975:or
610:SOM
600:GAN
576:ESN
571:GRU
516:-NN
451:SDL
441:PGD
436:PCA
431:NMF
426:LDA
421:ICA
416:CCA
292:-NN
2182::
2137:.
2129:.
2119:37
2117:.
2092:.
2069:.
2061:.
2053:.
2043:62
2041:.
2016:.
1992:.
1967:^
1953:.
1943:.
1929:.
1925:.
1902:.
1892:.
1852:.
1840:.
1836:.
1806:.
1781:.
1767:.
1763:.
1740:.
1728:.
1705:.
1695:45
1693:.
1670:.
1660:.
1652:.
1638:.
1608:^
1594:.
1584:.
1552:.
1542:.
1534:.
1482:^
1367:,
1175:.
1029:,
1006:.
979:.
879:ML
2168:.
2145:.
2125::
2102:.
2096::
2077:.
2057::
2049::
2026:.
2002:.
1988::
1961:.
1947::
1937::
1931:5
1888::
1860:.
1848::
1818:.
1791:.
1785::
1775::
1769:2
1748:.
1736::
1730:3
1713:.
1709::
1701::
1678:.
1656::
1602:.
1580::
1560:.
1538::
1530::
1506:.
1406:W
1402:W
1395:T
1391:W
1384:n
1380:W
1373:T
1369:W
1265:.
1170:T
1155:T
1139:i
1136:,
1133:C
1128:T
1116:.
1098:i
1095:,
1092:U
1087:T
1075:.
1057:i
1054:,
1051:K
1046:T
1031:T
1027:i
1020:T
1016:T
948:e
941:t
934:v
514:k
363:k
290:k
248:)
236:(
23:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.