Knowledge (XXG)

Bhattacharyya distance

Source 📝

3471: 3245: 2061: 3158: 3466:{\displaystyle D_{B}(p_{1},p_{2})={1 \over 8}({\boldsymbol {\mu }}_{1}-{\boldsymbol {\mu }}_{2})^{T}{\boldsymbol {\Sigma }}^{-1}({\boldsymbol {\mu }}_{1}-{\boldsymbol {\mu }}_{2})+{1 \over 2}\ln \,\left({\det {\boldsymbol {\Sigma }} \over {\sqrt {\det {\boldsymbol {\Sigma }}_{1}\,\det {\boldsymbol {\Sigma }}_{2}}}}\right)} 85:. He has developed this through a series of papers. He developed the method to measure the distance between two non-normal distributions and illustrated this with the classical multinomial populations, this work despite being submitted for publication in 1941, appeared almost five years later in 4273:
is a special case of the Bhattacharyya distance when the two classes are normally distributed with the same variances. When two classes have similar means but significantly different variances, the Mahalanobis distance would be close to zero, while the Bhattacharyya distance would not be.
1285: 2940: 1690: 2700: 1658: 2212: 3539: 89:. Consequently, Professor Bhattacharyya started working toward developing a distance metric for probability distributions that are absolutely continuous with respect to the Lebesgue measure and published his progress in 1942, at Proceedings of the 3237: 3804: 4181: 1075: 2490: 356: 629: 3874: 3153:{\displaystyle D_{B}(p,q)={\frac {1}{4}}{\frac {(\mu _{p}-\mu _{q})^{2}}{\sigma _{p}^{2}+\sigma _{q}^{2}}}+{\frac {1}{2}}\ln \left({\frac {\sigma _{p}^{2}+\sigma _{q}^{2}}{2\sigma _{p}\sigma _{q}}}\right)} 2835: 2771: 2056:{\displaystyle bc(dx|P,Q)={\sqrt {p(x)q(x)}}\,\lambda (dx)={\sqrt {p(x)q(x)}}\,l(x)\mu (x)={\sqrt {p(x)l(x)q(x)\,l(x)}}\mu (dx)={\sqrt {p'(x)l'(x)q'(x)l'(x)}}\,\mu (dx)={\sqrt {p'(x)q'(x)}}\,\lambda '(dx)} 4738:
Djouadi, A.; Snorrason, O.; Garber, F. (1990). "The quality of Training-Sample estimates of the Bhattacharyya coefficient". IEEE Transactions on Pattern Analysis and Machine Intelligence. 12 (1): 92–97.
261: 2881: 1067: 699: 2485: 1459: 1027: 3934: 3624: 911: 849: 3708: 2324: 2073: 3480: 1411: 2386: 3727: 3585: 1373: 173: 476: 423: 2932: 2274: 1348: 1308: 977: 957: 787: 719: 4260: 4222: 2254: 4085: 4058: 3653: 2905: 1467: 1328: 534: 505: 3971: 3171: 937: 659: 4887: 4488: 4031: 4011: 3991: 2406: 1682: 763: 743: 146: 126: 4622:
François Goudail, Philippe Réfrégier, Guillaume Delyon, "Bhattacharyya distance as a contrast parameter for statistical processing of noisy optical images",
4735:
Kailath, T. (1967). "The Divergence and Bhattacharyya Distance Measures in Signal Selection". IEEE Transactions on Communication Technology. 15 (1): 52–60.
4093: 4565:
Arıkan, Erdal (July 2009). "Channel polarization: A method for constructing capacity-achieving codes for symmetric binary-input memoryless channels".
1280:{\displaystyle bc(dx|P,Q)={\sqrt {p(x)q(x)}}\,\lambda (dx)={\sqrt {{\frac {P(dx)}{\lambda (dx)}}(x){\frac {Q(dx)}{\lambda (dx)}}(x)}}\lambda (dx).} 4363: 272: 4732:
Nielsen, F.; Boltz, S. (2010). "The Burbea–Rao and Bhattacharyya centroids". IEEE Transactions on Information Theory. 57 (8): 5455–5466.
546: 90: 3811: 2776: 2712: 2695:{\displaystyle BC(P,Q)=\int _{\mathcal {X}}{\sqrt {p(x)}}Q(dx)=\int _{\mathcal {X}}{\sqrt {\frac {P(dx)}{Q(dx)}}}Q(dx)=E_{Q}\left} 4892: 4302: 4555:
Devroye, L., Gyorfi, L. & Lugosi, G. A Probabilistic Theory of Pattern Recognition. Discrete Appl Math 73, 192–194 (1997).
4882: 4715: 181: 94: 82: 4710: 2840: 1032: 664: 2411: 4332: 4278: 4266: 982: 4635:
Chang Huai You, "An SVM Kernel With GMM-Supervector Based on the Bhattacharyya Distance for Speaker Recognition",
3879: 766: 74: 43: 3590: 4705: 4652:, 1996. ICSLP 96. Proceedings., Fourth International Conference on, Vol 4, pp. 2005–2008 vol.4, 3−6 Oct 1996 854: 792: 4484:"On a measure of divergence between two statistical populations defined by their probability distributions" 2207:{\displaystyle BC(P,Q)=\int _{\mathcal {X}}bc(dx|P,Q)=\int _{\mathcal {X}}{\sqrt {p(x)q(x)}}\,\lambda (dx)} 3534:{\displaystyle {\boldsymbol {\Sigma }}={{\boldsymbol {\Sigma }}_{1}+{\boldsymbol {\Sigma }}_{2} \over 2}.} 3665: 2279: 1416: 1378: 4517:"The Perfect Marriage and Much More: Combining Dimension Reduction, Distance Measures and Covariance" 4312: 4270: 3542: 2329: 722: 149: 58: 3555: 154: 4297: 4285: 3656: 3165: 2884: 537: 62: 4483: 428: 375: 4789: 4763: 4592: 4574: 4528: 4439: 4403: 4307: 4224:. The quality of estimation depends on the choice of buckets; too few buckets would overestimate 3660: 2910: 39: 4461: 1353: 4828: 4781: 4683: 4359: 2259: 1333: 1293: 962: 942: 772: 704: 86: 4227: 4189: 2221: 1653:{\displaystyle P(dx)=p(x)\lambda (dx)=p'(x)\lambda '(dx)=p(x)l(x)\mu (dx)=p'(x)l'(x)\mu (dx)} 4859: 4820: 4773: 4675: 4584: 4538: 4395: 4383: 4322: 3944:
The Bhattacharyya coefficient quantifies the "closeness" of two random statistical samples.
3719: 3232:{\displaystyle p_{i}={\mathcal {N}}({\boldsymbol {\mu }}_{i},\,{\boldsymbol {\Sigma }}_{i})} 4501: 4063: 4036: 3631: 4497: 4284:
The Bhattacharyya distance is used in feature extraction and selection, image processing,
2890: 1313: 510: 481: 3950: 979:-almost everywhere. Such a measure, even such a probability measure, always exists, e.g. 916: 638: 17: 4317: 4016: 3996: 3976: 3799:{\displaystyle {\frac {1}{2}}-{\frac {1}{2}}{\sqrt {1-4\rho ^{2}}}\leq L^{*}\leq \rho } 2391: 1667: 748: 728: 131: 111: 4269:
is estimating the separability of classes. Up to a multiplicative factor, the squared
4876: 4407: 4793: 4609:
Euisun Choi, Chulhee Lee, "Feature extraction based on the Bhattacharyya distance",
4327: 78: 4727: 4596: 4176:{\displaystyle BC(\mathbf {p} ,\mathbf {q} )=\sum _{i=1}^{n}{\sqrt {p_{i}q_{i}}},} 73:
Both the Bhattacharyya distance and the Bhattacharyya coefficient are named after
4662:
Chattopadhyay, Aparna; Chattopadhyay, Asis Kumar; B-Rao, Chandrika (2004-06-01).
4353: 4808: 4751: 4542: 4847: 4824: 4664:"Bhattacharyya's distance measure as a precursor of genetic distance measures" 51: 31: 4832: 4785: 4777: 4687: 4588: 4399: 4848:"The quality of training sample estimates of the Bhattacharyya coefficient" 4722: 4809:"The Divergence and Bhattacharyya Distance Measures in Signal Selection" 4443: 4427: 4679: 4863: 4663: 351:{\displaystyle BC(P,Q)=\sum _{x\in {\mathcal {X}}}{\sqrt {P(x)Q(x)}}} 3718:
The Bhattacharyya distance can be used to upper and lower bound the
4533: 4768: 4579: 4516: 624:{\displaystyle BC(P,Q)=\int _{\mathcal {X}}{\sqrt {p(x)q(x)}}\,dx} 4428:"On a Measure of Divergence between Two Multinomial Populations" 3869:{\displaystyle \rho =\mathbb {E} {\sqrt {\eta (X)(1-\eta (X))}}} 93:
and the final work has appeared in 1943 in the Bulletin of the
61:, despite being named a "distance", since it does not obey the 4852:
IEEE Transactions on Pattern Analysis and Machine Intelligence
4648:
Mak, B., "Phone clustering using the Bhattacharyya distance",
4384:"Anil Kumar Bhattacharyya (1915-1996): A Reverent Remembrance" 2830:{\displaystyle q\sim {\mathcal {N}}(\mu _{q},\sigma _{q}^{2})} 2766:{\displaystyle p\sim {\mathcal {N}}(\mu _{p},\sigma _{p}^{2})} 4277:
The Bhattacharyya coefficient is used in the construction of
3190: 2846: 2788: 2724: 2570: 2524: 2154: 2107: 1051: 1041: 683: 673: 580: 313: 160: 50:, which is a measure of the amount of overlap between two 540:
functions, the Bhattacharyya coefficient is defined as
993: 4521:
Physica A: Statistical Mechanics and its Applications
4230: 4192: 4096: 4066: 4039: 4019: 3999: 3979: 3953: 3882: 3814: 3730: 3668: 3634: 3593: 3558: 3483: 3248: 3174: 2943: 2913: 2893: 2843: 2779: 2715: 2493: 2414: 2394: 2332: 2282: 2262: 2224: 2076: 1693: 1670: 1470: 1419: 1381: 1356: 1336: 1316: 1296: 1078: 1035: 985: 965: 945: 919: 857: 795: 775: 751: 731: 707: 667: 641: 549: 513: 484: 431: 378: 275: 184: 157: 134: 114: 256:{\displaystyle D_{B}(P,Q)=-\ln \left(BC(P,Q)\right)} 4613:, Volume 36, Issue 8, August 2003, Pages 1703–1709 4254: 4216: 4175: 4079: 4052: 4025: 4005: 3985: 3965: 3928: 3868: 3798: 3702: 3647: 3618: 3579: 3533: 3465: 3231: 3152: 2926: 2899: 2875: 2829: 2765: 2694: 2479: 2400: 2380: 2318: 2268: 2248: 2206: 2055: 1676: 1652: 1453: 1405: 1367: 1342: 1322: 1302: 1279: 1061: 1021: 971: 951: 931: 905: 843: 781: 757: 737: 713: 693: 653: 623: 528: 499: 470: 417: 350: 255: 167: 140: 120: 4846:Djouadi, A.; Snorrason, O.; Garber, F.D. (1990). 4723:Statistical Intuition of Bhattacharyya's distance 4728:Some of the properties of Bhattacharyya Distance 3440: 3424: 3413: 2876:{\displaystyle {\mathcal {N}}(\mu ,\sigma ^{2})} 2067:We finally define the Bhattacharyya coefficient 4087:, then the sample Bhattacharyya coefficient is 3993:buckets, and let the frequency of samples from 1062:{\displaystyle ({\mathcal {X}},{\mathcal {B}})} 694:{\displaystyle ({\mathcal {X}},{\mathcal {B}})} 635:More generally, given two probability measures 2480:{\displaystyle p(x)={\frac {P(dx)}{Q(dx)}}(x)} 4489:Bulletin of the Calcutta Mathematical Society 1022:{\displaystyle \lambda ={\tfrac {1}{2}}(P+Q)} 8: 4752:"The Burbea-Rao and Bhattacharyya Centroids" 3929:{\displaystyle \eta (X)=\mathbb {P} (Y=1|X)} 1029:. Then define the Bhattacharyya measure on 175:, the Bhattacharyya distance is defined as 38:is a quantity which represents a notion of 27:Similarity of two probability distributions 4466:Proceedings of the Indian Science Congress 4355:The Oxford Dictionary of Statistical Terms 4767: 4578: 4532: 4388:Calcutta Statistical Association Bulletin 4229: 4191: 4162: 4152: 4146: 4140: 4129: 4114: 4106: 4095: 4071: 4065: 4044: 4038: 4018: 3998: 3978: 3952: 3915: 3899: 3898: 3881: 3826: 3822: 3821: 3813: 3784: 3769: 3754: 3744: 3731: 3729: 3669: 3667: 3639: 3633: 3604: 3592: 3557: 3516: 3511: 3501: 3496: 3492: 3484: 3482: 3449: 3444: 3439: 3433: 3428: 3422: 3416: 3410: 3405: 3392: 3380: 3375: 3365: 3360: 3347: 3342: 3335: 3325: 3320: 3310: 3305: 3291: 3279: 3266: 3253: 3247: 3220: 3215: 3213: 3204: 3199: 3189: 3188: 3179: 3173: 3137: 3127: 3112: 3107: 3094: 3089: 3082: 3062: 3050: 3045: 3032: 3027: 3015: 3005: 2992: 2982: 2972: 2948: 2942: 2918: 2912: 2892: 2864: 2845: 2844: 2842: 2818: 2813: 2800: 2787: 2786: 2778: 2754: 2749: 2736: 2723: 2722: 2714: 2648: 2638: 2576: 2569: 2568: 2530: 2523: 2522: 2492: 2430: 2413: 2393: 2331: 2281: 2261: 2223: 2188: 2160: 2153: 2152: 2128: 2106: 2105: 2075: 2032: 1994: 1975: 1903: 1870: 1832: 1804: 1776: 1757: 1729: 1709: 1692: 1669: 1469: 1418: 1380: 1355: 1335: 1315: 1295: 1210: 1163: 1161: 1142: 1114: 1094: 1077: 1050: 1049: 1040: 1039: 1034: 992: 984: 964: 944: 918: 856: 794: 774: 750: 730: 706: 682: 681: 672: 671: 666: 640: 614: 586: 579: 578: 548: 512: 483: 430: 377: 320: 312: 311: 304: 274: 189: 183: 159: 158: 156: 133: 113: 4626:, Vol. 21, Issue 7, pp. 1231−1240 (2004) 4756:IEEE Transactions on Information Theory 4750:Nielsen, Frank; Boltz, Sylvain (2011). 4742: 4567:IEEE Transactions on Information Theory 4344: 3947:Given two sequences from distributions 3619:{\displaystyle 0\leq D_{B}\leq \infty } 3376: 3361: 3321: 3306: 3200: 4262:, while too many would underestimate. 3541:Note that the first term is a squared 906:{\displaystyle Q(dx)=q(x)\lambda (dx)} 844:{\displaystyle P(dx)=p(x)\lambda (dx)} 4288:, phone clustering, and in genetics. 361:is the Bhattacharyya coefficient for 7: 4888:Statistical deviation and dispersion 4477: 4475: 4455: 4453: 4421: 4419: 4417: 4377: 4375: 370:continuous probability distributions 4813:IEEE Transactions on Communications 3703:{\displaystyle {\sqrt {1-BC(p,q)}}} 2319:{\displaystyle 0\leq BC(P,Q)\leq 1} 1454:{\displaystyle \lambda '=l'(x)\mu } 4462:"On discrimination and divergence" 3613: 1290:It does not depend on the measure 913:for probability density functions 363:discrete probability distributions 25: 1406:{\displaystyle \lambda =l(x)\mu } 4482:Bhattacharyya, A. (March 1943). 4115: 4107: 3512: 3497: 3485: 3445: 3429: 3417: 3343: 3216: 2388:is absolutely continuous wrt to 4639:, IEEE, Vol 16, Is 1, pp. 49-52 2381:{\displaystyle P(dx)=p(x)Q(dx)} 2276:, and by the Cauchy inequality 1375:are absolutely continuous i.e. 81:who worked in the 1930s at the 46:. It is closely related to the 4249: 4237: 4211: 4199: 4119: 4103: 3936:is the posterior probability. 3923: 3916: 3903: 3892: 3886: 3861: 3858: 3852: 3840: 3837: 3831: 3695: 3683: 3580:{\displaystyle 0\leq BC\leq 1} 3386: 3356: 3332: 3301: 3285: 3259: 3226: 3195: 3012: 2985: 2966: 2954: 2870: 2851: 2824: 2793: 2760: 2729: 2681: 2672: 2664: 2655: 2628: 2619: 2609: 2600: 2592: 2583: 2558: 2549: 2541: 2535: 2512: 2500: 2474: 2468: 2462: 2453: 2445: 2436: 2424: 2418: 2408:with Radon Nikodym derivative 2375: 2366: 2360: 2354: 2345: 2336: 2307: 2295: 2243: 2231: 2201: 2192: 2183: 2177: 2171: 2165: 2142: 2129: 2119: 2095: 2083: 2050: 2041: 2027: 2021: 2010: 2004: 1988: 1979: 1970: 1964: 1953: 1947: 1936: 1930: 1919: 1913: 1897: 1888: 1880: 1874: 1867: 1861: 1855: 1849: 1843: 1837: 1826: 1820: 1814: 1808: 1799: 1793: 1787: 1781: 1770: 1761: 1752: 1746: 1740: 1734: 1723: 1710: 1700: 1647: 1638: 1632: 1626: 1615: 1609: 1595: 1586: 1580: 1574: 1568: 1562: 1553: 1544: 1533: 1527: 1513: 1504: 1498: 1492: 1483: 1474: 1445: 1439: 1397: 1391: 1271: 1262: 1254: 1248: 1242: 1233: 1225: 1216: 1207: 1201: 1195: 1186: 1178: 1169: 1155: 1146: 1137: 1131: 1125: 1119: 1108: 1095: 1085: 1056: 1036: 1016: 1004: 900: 891: 885: 879: 870: 861: 838: 829: 823: 817: 808: 799: 688: 668: 609: 603: 597: 591: 568: 556: 523: 517: 494: 488: 459: 453: 444: 435: 406: 400: 391: 382: 343: 337: 331: 325: 294: 282: 245: 233: 207: 195: 168:{\displaystyle {\mathcal {X}}} 1: 1310:, for if we choose a measure 95:Calcutta Mathematical Society 4468:. Asiatic Society of Bengal. 1350:and an other measure choice 471:{\displaystyle Q(dx)=q(x)dx} 418:{\displaystyle P(dx)=p(x)dx} 83:Indian Statistical Institute 4711:Encyclopedia of Mathematics 4543:10.1016/j.physa.2019.04.174 4358:. Oxford University Press. 4303:Kullback–Leibler divergence 2927:{\displaystyle \sigma ^{2}} 2218:By the above, the quantity 4909: 4426:Bhattacharyya, A. (1946). 4382:Sen, Pranab Kumar (1996). 4333:Fidelity of quantum states 3164:And in general, given two 4825:10.1109/TCOM.1967.1089532 4637:Signal Processing Letters 4460:Bhattacharyya, A (1942). 4186:which is an estimator of 1368:{\displaystyle \lambda '} 107:probability distributions 48:Bhattacharyya coefficient 44:probability distributions 18:Bhattacharyya coefficient 4778:10.1109/TIT.2011.2159046 4706:"Bhattacharyya distance" 4589:10.1109/TIT.2009.2021379 4400:10.1177/0008068319960301 2269:{\displaystyle \lambda } 1343:{\displaystyle \lambda } 1303:{\displaystyle \lambda } 972:{\displaystyle \lambda } 952:{\displaystyle \lambda } 782:{\displaystyle \lambda } 714:{\displaystyle \lambda } 75:Anil Kumar Bhattacharyya 54:samples or populations. 4893:Anil Kumar Bhattacharya 4352:Dodge, Yadolah (2003). 4255:{\displaystyle BC(P,Q)} 4217:{\displaystyle BC(P,Q)} 2249:{\displaystyle BC(P,Q)} 91:Indian Science Congress 4668:Journal of Biosciences 4515:Kashyap, Ravi (2019). 4256: 4218: 4177: 4145: 4081: 4054: 4027: 4007: 3987: 3967: 3930: 3870: 3800: 3704: 3649: 3620: 3581: 3535: 3467: 3233: 3154: 2928: 2901: 2877: 2831: 2767: 2696: 2481: 2402: 2382: 2320: 2270: 2250: 2208: 2057: 1678: 1654: 1455: 1407: 1369: 1344: 1324: 1304: 1281: 1063: 1023: 973: 953: 933: 907: 845: 783: 759: 739: 715: 695: 661:on a measurable space 655: 625: 530: 501: 472: 419: 352: 257: 169: 142: 122: 36:Bhattacharyya distance 4257: 4219: 4178: 4125: 4082: 4080:{\displaystyle q_{i}} 4055: 4053:{\displaystyle p_{i}} 4028: 4008: 3988: 3968: 3931: 3871: 3801: 3714:Bounds on Bayes error 3705: 3650: 3648:{\displaystyle D_{B}} 3621: 3582: 3536: 3468: 3234: 3155: 2929: 2902: 2878: 2832: 2768: 2697: 2482: 2403: 2383: 2321: 2271: 2251: 2209: 2058: 1679: 1655: 1456: 1408: 1370: 1345: 1325: 1305: 1282: 1064: 1024: 974: 954: 934: 908: 846: 784: 767:absolutely continuous 760: 740: 716: 696: 656: 626: 531: 502: 473: 420: 353: 258: 170: 143: 123: 4883:Statistical distance 4807:Kailath, T. (1967). 4313:Mahalanobis distance 4271:Mahalanobis distance 4228: 4190: 4094: 4064: 4060:, and similarly for 4037: 4017: 3997: 3977: 3951: 3880: 3812: 3728: 3666: 3632: 3591: 3556: 3543:Mahalanobis distance 3481: 3246: 3172: 2941: 2911: 2900:{\displaystyle \mu } 2891: 2841: 2777: 2713: 2491: 2412: 2392: 2330: 2280: 2260: 2222: 2074: 1691: 1668: 1468: 1417: 1379: 1354: 1334: 1323:{\displaystyle \mu } 1314: 1294: 1076: 1033: 983: 963: 943: 917: 855: 793: 773: 749: 729: 725:) measure such that 705: 665: 639: 547: 529:{\displaystyle q(x)} 511: 500:{\displaystyle p(x)} 482: 429: 376: 273: 182: 155: 132: 112: 4611:Pattern Recognition 4298:Bhattacharyya angle 4286:speaker recognition 3966:{\displaystyle P,Q} 3657:triangle inequality 3166:multivariate normal 3117: 3099: 3055: 3037: 2885:normal distribution 2823: 2759: 2326:. In particular if 2256:does not depend on 932:{\displaystyle p,q} 654:{\displaystyle P,Q} 538:probability density 63:triangle inequality 4680:10.1007/BF02703410 4308:Hellinger distance 4252: 4214: 4173: 4077: 4050: 4023: 4003: 3983: 3963: 3926: 3866: 3796: 3700: 3661:Hellinger distance 3655:does not obey the 3645: 3616: 3577: 3531: 3463: 3229: 3150: 3103: 3085: 3041: 3023: 2924: 2897: 2873: 2827: 2809: 2763: 2745: 2692: 2477: 2398: 2378: 2316: 2266: 2246: 2204: 2053: 1674: 1664:and similarly for 1650: 1451: 1403: 1365: 1340: 1320: 1300: 1277: 1059: 1019: 1002: 969: 949: 929: 903: 841: 779: 755: 735: 711: 691: 651: 621: 526: 497: 468: 415: 348: 319: 253: 165: 138: 118: 4365:978-0-19-920613-1 4265:A common task in 4168: 4026:{\displaystyle i} 4006:{\displaystyle P} 3986:{\displaystyle n} 3864: 3775: 3752: 3739: 3698: 3526: 3457: 3455: 3400: 3299: 3144: 3070: 3057: 2980: 2686: 2685: 2614: 2613: 2544: 2466: 2401:{\displaystyle Q} 2186: 2030: 1973: 1883: 1802: 1755: 1677:{\displaystyle Q} 1257: 1246: 1199: 1140: 1001: 758:{\displaystyle Q} 738:{\displaystyle P} 612: 346: 300: 141:{\displaystyle Q} 121:{\displaystyle P} 16:(Redirected from 4900: 4868: 4867: 4864:10.1109/34.41388 4843: 4837: 4836: 4804: 4798: 4797: 4771: 4762:(8): 5455–5466. 4747: 4719: 4692: 4691: 4659: 4653: 4646: 4640: 4633: 4627: 4620: 4614: 4607: 4601: 4600: 4582: 4573:(7): 3051–3073. 4562: 4556: 4553: 4547: 4546: 4536: 4512: 4506: 4505: 4479: 4470: 4469: 4457: 4448: 4447: 4423: 4412: 4411: 4394:(3–4): 151–158. 4379: 4370: 4369: 4349: 4261: 4259: 4258: 4253: 4223: 4221: 4220: 4215: 4182: 4180: 4179: 4174: 4169: 4167: 4166: 4157: 4156: 4147: 4144: 4139: 4118: 4110: 4086: 4084: 4083: 4078: 4076: 4075: 4059: 4057: 4056: 4051: 4049: 4048: 4032: 4030: 4029: 4024: 4012: 4010: 4009: 4004: 3992: 3990: 3989: 3984: 3973:, bin them into 3972: 3970: 3969: 3964: 3935: 3933: 3932: 3927: 3919: 3902: 3875: 3873: 3872: 3867: 3865: 3827: 3825: 3805: 3803: 3802: 3797: 3789: 3788: 3776: 3774: 3773: 3755: 3753: 3745: 3740: 3732: 3720:Bayes error rate 3709: 3707: 3706: 3701: 3699: 3670: 3654: 3652: 3651: 3646: 3644: 3643: 3625: 3623: 3622: 3617: 3609: 3608: 3586: 3584: 3583: 3578: 3540: 3538: 3537: 3532: 3527: 3522: 3521: 3520: 3515: 3506: 3505: 3500: 3493: 3488: 3472: 3470: 3469: 3464: 3462: 3458: 3456: 3454: 3453: 3448: 3438: 3437: 3432: 3423: 3421: 3420: 3411: 3401: 3393: 3385: 3384: 3379: 3370: 3369: 3364: 3355: 3354: 3346: 3340: 3339: 3330: 3329: 3324: 3315: 3314: 3309: 3300: 3292: 3284: 3283: 3271: 3270: 3258: 3257: 3238: 3236: 3235: 3230: 3225: 3224: 3219: 3209: 3208: 3203: 3194: 3193: 3184: 3183: 3159: 3157: 3156: 3151: 3149: 3145: 3143: 3142: 3141: 3132: 3131: 3118: 3116: 3111: 3098: 3093: 3083: 3071: 3063: 3058: 3056: 3054: 3049: 3036: 3031: 3021: 3020: 3019: 3010: 3009: 2997: 2996: 2983: 2981: 2973: 2953: 2952: 2933: 2931: 2930: 2925: 2923: 2922: 2906: 2904: 2903: 2898: 2882: 2880: 2879: 2874: 2869: 2868: 2850: 2849: 2836: 2834: 2833: 2828: 2822: 2817: 2805: 2804: 2792: 2791: 2772: 2770: 2769: 2764: 2758: 2753: 2741: 2740: 2728: 2727: 2701: 2699: 2698: 2693: 2691: 2687: 2684: 2667: 2650: 2649: 2643: 2642: 2615: 2612: 2595: 2578: 2577: 2575: 2574: 2573: 2545: 2531: 2529: 2528: 2527: 2486: 2484: 2483: 2478: 2467: 2465: 2448: 2431: 2407: 2405: 2404: 2399: 2387: 2385: 2384: 2379: 2325: 2323: 2322: 2317: 2275: 2273: 2272: 2267: 2255: 2253: 2252: 2247: 2213: 2211: 2210: 2205: 2187: 2161: 2159: 2158: 2157: 2132: 2112: 2111: 2110: 2062: 2060: 2059: 2054: 2040: 2031: 2020: 2003: 1995: 1974: 1963: 1946: 1929: 1912: 1904: 1884: 1833: 1803: 1777: 1756: 1730: 1713: 1683: 1681: 1680: 1675: 1659: 1657: 1656: 1651: 1625: 1608: 1543: 1526: 1460: 1458: 1457: 1452: 1438: 1427: 1412: 1410: 1409: 1404: 1374: 1372: 1371: 1366: 1364: 1349: 1347: 1346: 1341: 1329: 1327: 1326: 1321: 1309: 1307: 1306: 1301: 1286: 1284: 1283: 1278: 1258: 1247: 1245: 1228: 1211: 1200: 1198: 1181: 1164: 1162: 1141: 1115: 1098: 1068: 1066: 1065: 1060: 1055: 1054: 1045: 1044: 1028: 1026: 1025: 1020: 1003: 994: 978: 976: 975: 970: 958: 956: 955: 950: 939:with respect to 938: 936: 935: 930: 912: 910: 909: 904: 850: 848: 847: 842: 788: 786: 785: 780: 769:with respect to 764: 762: 761: 756: 744: 742: 741: 736: 720: 718: 717: 712: 700: 698: 697: 692: 687: 686: 677: 676: 660: 658: 657: 652: 630: 628: 627: 622: 613: 587: 585: 584: 583: 535: 533: 532: 527: 506: 504: 503: 498: 477: 475: 474: 469: 424: 422: 421: 416: 357: 355: 354: 349: 347: 321: 318: 317: 316: 262: 260: 259: 254: 252: 248: 194: 193: 174: 172: 171: 166: 164: 163: 147: 145: 144: 139: 127: 125: 124: 119: 21: 4908: 4907: 4903: 4902: 4901: 4899: 4898: 4897: 4873: 4872: 4871: 4845: 4844: 4840: 4806: 4805: 4801: 4749: 4748: 4744: 4704: 4701: 4696: 4695: 4661: 4660: 4656: 4650:Spoken Language 4647: 4643: 4634: 4630: 4621: 4617: 4608: 4604: 4564: 4563: 4559: 4554: 4550: 4514: 4513: 4509: 4481: 4480: 4473: 4459: 4458: 4451: 4425: 4424: 4415: 4381: 4380: 4373: 4366: 4351: 4350: 4346: 4341: 4294: 4226: 4225: 4188: 4187: 4158: 4148: 4092: 4091: 4067: 4062: 4061: 4040: 4035: 4034: 4015: 4014: 3995: 3994: 3975: 3974: 3949: 3948: 3942: 3878: 3877: 3810: 3809: 3780: 3765: 3726: 3725: 3716: 3664: 3663: 3635: 3630: 3629: 3600: 3589: 3588: 3554: 3553: 3551: 3510: 3495: 3494: 3479: 3478: 3443: 3427: 3412: 3406: 3374: 3359: 3341: 3331: 3319: 3304: 3275: 3262: 3249: 3244: 3243: 3214: 3198: 3175: 3170: 3169: 3133: 3123: 3119: 3084: 3078: 3022: 3011: 3001: 2988: 2984: 2944: 2939: 2938: 2914: 2909: 2908: 2889: 2888: 2860: 2839: 2838: 2796: 2775: 2774: 2732: 2711: 2710: 2707: 2668: 2651: 2644: 2634: 2596: 2579: 2564: 2518: 2489: 2488: 2449: 2432: 2410: 2409: 2390: 2389: 2328: 2327: 2278: 2277: 2258: 2257: 2220: 2219: 2148: 2101: 2072: 2071: 2033: 2013: 1996: 1956: 1939: 1922: 1905: 1689: 1688: 1684:. We then have 1666: 1665: 1618: 1601: 1536: 1519: 1466: 1465: 1431: 1420: 1415: 1414: 1377: 1376: 1357: 1352: 1351: 1332: 1331: 1312: 1311: 1292: 1291: 1229: 1212: 1182: 1165: 1074: 1073: 1031: 1030: 981: 980: 961: 960: 941: 940: 915: 914: 853: 852: 791: 790: 789:i.e. such that 771: 770: 747: 746: 727: 726: 703: 702: 663: 662: 637: 636: 574: 545: 544: 509: 508: 480: 479: 427: 426: 374: 373: 271: 270: 226: 222: 185: 180: 179: 153: 152: 130: 129: 110: 109: 103: 71: 28: 23: 22: 15: 12: 11: 5: 4906: 4904: 4896: 4895: 4890: 4885: 4875: 4874: 4870: 4869: 4838: 4799: 4741: 4740: 4739: 4736: 4733: 4730: 4725: 4720: 4700: 4699:External links 4697: 4694: 4693: 4674:(2): 135–138. 4654: 4641: 4628: 4615: 4602: 4557: 4548: 4507: 4471: 4449: 4438:(4): 401–406. 4413: 4371: 4364: 4343: 4342: 4340: 4337: 4336: 4335: 4330: 4325: 4320: 4318:Chernoff bound 4315: 4310: 4305: 4300: 4293: 4290: 4267:classification 4251: 4248: 4245: 4242: 4239: 4236: 4233: 4213: 4210: 4207: 4204: 4201: 4198: 4195: 4184: 4183: 4172: 4165: 4161: 4155: 4151: 4143: 4138: 4135: 4132: 4128: 4124: 4121: 4117: 4113: 4109: 4105: 4102: 4099: 4074: 4070: 4047: 4043: 4022: 4002: 3982: 3962: 3959: 3956: 3941: 3938: 3925: 3922: 3918: 3914: 3911: 3908: 3905: 3901: 3897: 3894: 3891: 3888: 3885: 3863: 3860: 3857: 3854: 3851: 3848: 3845: 3842: 3839: 3836: 3833: 3830: 3824: 3820: 3817: 3795: 3792: 3787: 3783: 3779: 3772: 3768: 3764: 3761: 3758: 3751: 3748: 3743: 3738: 3735: 3715: 3712: 3697: 3694: 3691: 3688: 3685: 3682: 3679: 3676: 3673: 3642: 3638: 3615: 3612: 3607: 3603: 3599: 3596: 3576: 3573: 3570: 3567: 3564: 3561: 3550: 3547: 3530: 3525: 3519: 3514: 3509: 3504: 3499: 3491: 3487: 3475: 3474: 3461: 3452: 3447: 3442: 3436: 3431: 3426: 3419: 3415: 3409: 3404: 3399: 3396: 3391: 3388: 3383: 3378: 3373: 3368: 3363: 3358: 3353: 3350: 3345: 3338: 3334: 3328: 3323: 3318: 3313: 3308: 3303: 3298: 3295: 3290: 3287: 3282: 3278: 3274: 3269: 3265: 3261: 3256: 3252: 3228: 3223: 3218: 3212: 3207: 3202: 3197: 3192: 3187: 3182: 3178: 3168:distributions 3162: 3161: 3148: 3140: 3136: 3130: 3126: 3122: 3115: 3110: 3106: 3102: 3097: 3092: 3088: 3081: 3077: 3074: 3069: 3066: 3061: 3053: 3048: 3044: 3040: 3035: 3030: 3026: 3018: 3014: 3008: 3004: 3000: 2995: 2991: 2987: 2979: 2976: 2971: 2968: 2965: 2962: 2959: 2956: 2951: 2947: 2921: 2917: 2896: 2872: 2867: 2863: 2859: 2856: 2853: 2848: 2826: 2821: 2816: 2812: 2808: 2803: 2799: 2795: 2790: 2785: 2782: 2762: 2757: 2752: 2748: 2744: 2739: 2735: 2731: 2726: 2721: 2718: 2706: 2703: 2690: 2683: 2680: 2677: 2674: 2671: 2666: 2663: 2660: 2657: 2654: 2647: 2641: 2637: 2633: 2630: 2627: 2624: 2621: 2618: 2611: 2608: 2605: 2602: 2599: 2594: 2591: 2588: 2585: 2582: 2572: 2567: 2563: 2560: 2557: 2554: 2551: 2548: 2543: 2540: 2537: 2534: 2526: 2521: 2517: 2514: 2511: 2508: 2505: 2502: 2499: 2496: 2476: 2473: 2470: 2464: 2461: 2458: 2455: 2452: 2447: 2444: 2441: 2438: 2435: 2429: 2426: 2423: 2420: 2417: 2397: 2377: 2374: 2371: 2368: 2365: 2362: 2359: 2356: 2353: 2350: 2347: 2344: 2341: 2338: 2335: 2315: 2312: 2309: 2306: 2303: 2300: 2297: 2294: 2291: 2288: 2285: 2265: 2245: 2242: 2239: 2236: 2233: 2230: 2227: 2216: 2215: 2203: 2200: 2197: 2194: 2191: 2185: 2182: 2179: 2176: 2173: 2170: 2167: 2164: 2156: 2151: 2147: 2144: 2141: 2138: 2135: 2131: 2127: 2124: 2121: 2118: 2115: 2109: 2104: 2100: 2097: 2094: 2091: 2088: 2085: 2082: 2079: 2065: 2064: 2052: 2049: 2046: 2043: 2039: 2036: 2029: 2026: 2023: 2019: 2016: 2012: 2009: 2006: 2002: 1999: 1993: 1990: 1987: 1984: 1981: 1978: 1972: 1969: 1966: 1962: 1959: 1955: 1952: 1949: 1945: 1942: 1938: 1935: 1932: 1928: 1925: 1921: 1918: 1915: 1911: 1908: 1902: 1899: 1896: 1893: 1890: 1887: 1882: 1879: 1876: 1873: 1869: 1866: 1863: 1860: 1857: 1854: 1851: 1848: 1845: 1842: 1839: 1836: 1831: 1828: 1825: 1822: 1819: 1816: 1813: 1810: 1807: 1801: 1798: 1795: 1792: 1789: 1786: 1783: 1780: 1775: 1772: 1769: 1766: 1763: 1760: 1754: 1751: 1748: 1745: 1742: 1739: 1736: 1733: 1728: 1725: 1722: 1719: 1716: 1712: 1708: 1705: 1702: 1699: 1696: 1673: 1662: 1661: 1649: 1646: 1643: 1640: 1637: 1634: 1631: 1628: 1624: 1621: 1617: 1614: 1611: 1607: 1604: 1600: 1597: 1594: 1591: 1588: 1585: 1582: 1579: 1576: 1573: 1570: 1567: 1564: 1561: 1558: 1555: 1552: 1549: 1546: 1542: 1539: 1535: 1532: 1529: 1525: 1522: 1518: 1515: 1512: 1509: 1506: 1503: 1500: 1497: 1494: 1491: 1488: 1485: 1482: 1479: 1476: 1473: 1450: 1447: 1444: 1441: 1437: 1434: 1430: 1426: 1423: 1402: 1399: 1396: 1393: 1390: 1387: 1384: 1363: 1360: 1339: 1319: 1299: 1288: 1287: 1276: 1273: 1270: 1267: 1264: 1261: 1256: 1253: 1250: 1244: 1241: 1238: 1235: 1232: 1227: 1224: 1221: 1218: 1215: 1209: 1206: 1203: 1197: 1194: 1191: 1188: 1185: 1180: 1177: 1174: 1171: 1168: 1160: 1157: 1154: 1151: 1148: 1145: 1139: 1136: 1133: 1130: 1127: 1124: 1121: 1118: 1113: 1110: 1107: 1104: 1101: 1097: 1093: 1090: 1087: 1084: 1081: 1058: 1053: 1048: 1043: 1038: 1018: 1015: 1012: 1009: 1006: 1000: 997: 991: 988: 968: 948: 928: 925: 922: 902: 899: 896: 893: 890: 887: 884: 881: 878: 875: 872: 869: 866: 863: 860: 840: 837: 834: 831: 828: 825: 822: 819: 816: 813: 810: 807: 804: 801: 798: 778: 754: 734: 710: 690: 685: 680: 675: 670: 650: 647: 644: 633: 632: 620: 617: 611: 608: 605: 602: 599: 596: 593: 590: 582: 577: 573: 570: 567: 564: 561: 558: 555: 552: 525: 522: 519: 516: 496: 493: 490: 487: 467: 464: 461: 458: 455: 452: 449: 446: 443: 440: 437: 434: 414: 411: 408: 405: 402: 399: 396: 393: 390: 387: 384: 381: 359: 358: 345: 342: 339: 336: 333: 330: 327: 324: 315: 310: 307: 303: 299: 296: 293: 290: 287: 284: 281: 278: 264: 263: 251: 247: 244: 241: 238: 235: 232: 229: 225: 221: 218: 215: 212: 209: 206: 203: 200: 197: 192: 188: 162: 137: 117: 102: 99: 70: 67: 26: 24: 14: 13: 10: 9: 6: 4: 3: 2: 4905: 4894: 4891: 4889: 4886: 4884: 4881: 4880: 4878: 4865: 4861: 4857: 4853: 4849: 4842: 4839: 4834: 4830: 4826: 4822: 4818: 4814: 4810: 4803: 4800: 4795: 4791: 4787: 4783: 4779: 4775: 4770: 4765: 4761: 4757: 4753: 4746: 4743: 4737: 4734: 4731: 4729: 4726: 4724: 4721: 4717: 4713: 4712: 4707: 4703: 4702: 4698: 4689: 4685: 4681: 4677: 4673: 4669: 4665: 4658: 4655: 4651: 4645: 4642: 4638: 4632: 4629: 4625: 4619: 4616: 4612: 4606: 4603: 4598: 4594: 4590: 4586: 4581: 4576: 4572: 4568: 4561: 4558: 4552: 4549: 4544: 4540: 4535: 4530: 4526: 4522: 4518: 4511: 4508: 4503: 4499: 4495: 4491: 4490: 4485: 4478: 4476: 4472: 4467: 4463: 4456: 4454: 4450: 4445: 4441: 4437: 4433: 4429: 4422: 4420: 4418: 4414: 4409: 4405: 4401: 4397: 4393: 4389: 4385: 4378: 4376: 4372: 4367: 4361: 4357: 4356: 4348: 4345: 4338: 4334: 4331: 4329: 4326: 4324: 4323:RĂ©nyi entropy 4321: 4319: 4316: 4314: 4311: 4309: 4306: 4304: 4301: 4299: 4296: 4295: 4291: 4289: 4287: 4282: 4280: 4275: 4272: 4268: 4263: 4246: 4243: 4240: 4234: 4231: 4208: 4205: 4202: 4196: 4193: 4170: 4163: 4159: 4153: 4149: 4141: 4136: 4133: 4130: 4126: 4122: 4111: 4100: 4097: 4090: 4089: 4088: 4072: 4068: 4045: 4041: 4020: 4000: 3980: 3960: 3957: 3954: 3945: 3939: 3937: 3920: 3912: 3909: 3906: 3895: 3889: 3883: 3855: 3849: 3846: 3843: 3834: 3828: 3818: 3815: 3806: 3793: 3790: 3785: 3781: 3777: 3770: 3766: 3762: 3759: 3756: 3749: 3746: 3741: 3736: 3733: 3723: 3721: 3713: 3711: 3692: 3689: 3686: 3680: 3677: 3674: 3671: 3662: 3659:, though the 3658: 3640: 3636: 3627: 3610: 3605: 3601: 3597: 3594: 3574: 3571: 3568: 3565: 3562: 3559: 3548: 3546: 3544: 3528: 3523: 3517: 3507: 3502: 3489: 3459: 3450: 3434: 3407: 3402: 3397: 3394: 3389: 3381: 3371: 3366: 3351: 3348: 3336: 3326: 3316: 3311: 3296: 3293: 3288: 3280: 3276: 3272: 3267: 3263: 3254: 3250: 3242: 3241: 3240: 3221: 3210: 3205: 3185: 3180: 3176: 3167: 3146: 3138: 3134: 3128: 3124: 3120: 3113: 3108: 3104: 3100: 3095: 3090: 3086: 3079: 3075: 3072: 3067: 3064: 3059: 3051: 3046: 3042: 3038: 3033: 3028: 3024: 3016: 3006: 3002: 2998: 2993: 2989: 2977: 2974: 2969: 2963: 2960: 2957: 2949: 2945: 2937: 2936: 2935: 2919: 2915: 2907:and variance 2894: 2886: 2865: 2861: 2857: 2854: 2819: 2814: 2810: 2806: 2801: 2797: 2783: 2780: 2755: 2750: 2746: 2742: 2737: 2733: 2719: 2716: 2705:Gaussian case 2704: 2702: 2688: 2678: 2675: 2669: 2661: 2658: 2652: 2645: 2639: 2635: 2631: 2625: 2622: 2616: 2606: 2603: 2597: 2589: 2586: 2580: 2565: 2561: 2555: 2552: 2546: 2538: 2532: 2519: 2515: 2509: 2506: 2503: 2497: 2494: 2471: 2459: 2456: 2450: 2442: 2439: 2433: 2427: 2421: 2415: 2395: 2372: 2369: 2363: 2357: 2351: 2348: 2342: 2339: 2333: 2313: 2310: 2304: 2301: 2298: 2292: 2289: 2286: 2283: 2263: 2240: 2237: 2234: 2228: 2225: 2198: 2195: 2189: 2180: 2174: 2168: 2162: 2149: 2145: 2139: 2136: 2133: 2125: 2122: 2116: 2113: 2102: 2098: 2092: 2089: 2086: 2080: 2077: 2070: 2069: 2068: 2047: 2044: 2037: 2034: 2024: 2017: 2014: 2007: 2000: 1997: 1991: 1985: 1982: 1976: 1967: 1960: 1957: 1950: 1943: 1940: 1933: 1926: 1923: 1916: 1909: 1906: 1900: 1894: 1891: 1885: 1877: 1871: 1864: 1858: 1852: 1846: 1840: 1834: 1829: 1823: 1817: 1811: 1805: 1796: 1790: 1784: 1778: 1773: 1767: 1764: 1758: 1749: 1743: 1737: 1731: 1726: 1720: 1717: 1714: 1706: 1703: 1697: 1694: 1687: 1686: 1685: 1671: 1644: 1641: 1635: 1629: 1622: 1619: 1612: 1605: 1602: 1598: 1592: 1589: 1583: 1577: 1571: 1565: 1559: 1556: 1550: 1547: 1540: 1537: 1530: 1523: 1520: 1516: 1510: 1507: 1501: 1495: 1489: 1486: 1480: 1477: 1471: 1464: 1463: 1462: 1448: 1442: 1435: 1432: 1428: 1424: 1421: 1400: 1394: 1388: 1385: 1382: 1361: 1358: 1337: 1317: 1297: 1274: 1268: 1265: 1259: 1251: 1239: 1236: 1230: 1222: 1219: 1213: 1204: 1192: 1189: 1183: 1175: 1172: 1166: 1158: 1152: 1149: 1143: 1134: 1128: 1122: 1116: 1111: 1105: 1102: 1099: 1091: 1088: 1082: 1079: 1072: 1071: 1070: 1046: 1013: 1010: 1007: 998: 995: 989: 986: 966: 946: 926: 923: 920: 897: 894: 888: 882: 876: 873: 867: 864: 858: 835: 832: 826: 820: 814: 811: 805: 802: 796: 776: 768: 752: 732: 724: 708: 678: 648: 645: 642: 618: 615: 606: 600: 594: 588: 575: 571: 565: 562: 559: 553: 550: 543: 542: 541: 539: 520: 514: 491: 485: 465: 462: 456: 450: 447: 441: 438: 432: 412: 409: 403: 397: 394: 388: 385: 379: 371: 366: 364: 340: 334: 328: 322: 308: 305: 301: 297: 291: 288: 285: 279: 276: 269: 268: 267: 249: 242: 239: 236: 230: 227: 223: 219: 216: 213: 210: 204: 201: 198: 190: 186: 178: 177: 176: 151: 135: 115: 108: 100: 98: 96: 92: 88: 84: 80: 76: 68: 66: 64: 60: 55: 53: 49: 45: 41: 37: 33: 19: 4858:(1): 92–97. 4855: 4851: 4841: 4819:(1): 52–60. 4816: 4812: 4802: 4759: 4755: 4745: 4709: 4671: 4667: 4657: 4649: 4644: 4636: 4631: 4623: 4618: 4610: 4605: 4570: 4566: 4560: 4551: 4524: 4520: 4510: 4493: 4487: 4465: 4435: 4431: 4391: 4387: 4354: 4347: 4328:F-divergence 4283: 4276: 4264: 4185: 3946: 3943: 3940:Applications 3807: 3724: 3717: 3628: 3552: 3476: 3163: 2708: 2217: 2066: 1663: 1289: 723:sigma finite 634: 369: 367: 362: 360: 265: 148:on the same 106: 104: 79:statistician 72: 57:It is not a 56: 47: 42:between two 35: 29: 4279:polar codes 52:statistical 4877:Categories 4534:1603.09060 4527:: 120938. 4496:: 99–109. 4339:References 4013:in bucket 3549:Properties 2887:with mean 1330:such that 101:Definition 40:similarity 32:statistics 4833:0096-2244 4786:0018-9448 4769:1004.5049 4716:EMS Press 4688:0973-7138 4580:0807.3917 4408:164326977 4127:∑ 3884:η 3850:η 3847:− 3829:η 3816:ρ 3794:ρ 3791:≤ 3786:∗ 3778:≤ 3767:ρ 3760:− 3742:− 3675:− 3614:∞ 3611:≤ 3598:≤ 3572:≤ 3563:≤ 3513:Σ 3498:Σ 3486:Σ 3446:Σ 3430:Σ 3418:Σ 3377:μ 3372:− 3362:μ 3349:− 3344:Σ 3322:μ 3317:− 3307:μ 3217:Σ 3201:μ 3135:σ 3125:σ 3105:σ 3087:σ 3076:⁡ 3043:σ 3025:σ 3003:μ 2999:− 2990:μ 2916:σ 2895:μ 2862:σ 2855:μ 2811:σ 2798:μ 2784:∼ 2747:σ 2734:μ 2720:∼ 2566:∫ 2520:∫ 2311:≤ 2287:≤ 2264:λ 2190:λ 2150:∫ 2103:∫ 2035:λ 1977:μ 1886:μ 1818:μ 1759:λ 1636:μ 1584:μ 1538:λ 1502:λ 1449:μ 1422:λ 1401:μ 1383:λ 1359:λ 1338:λ 1318:μ 1298:λ 1260:λ 1231:λ 1184:λ 1144:λ 987:λ 967:λ 947:λ 889:λ 827:λ 777:λ 709:λ 576:∫ 309:∈ 302:∑ 220:⁡ 214:− 4794:14238708 4444:25047882 4292:See also 2837:, where 2038:′ 2018:′ 2001:′ 1961:′ 1944:′ 1927:′ 1910:′ 1623:′ 1606:′ 1541:′ 1524:′ 1461:, then 1436:′ 1425:′ 1362:′ 959:defined 536:are the 4718:, 2001 4502:0010358 4432:Sankhyā 2934:; then 2883:is the 2487:, then 372:, with 87:Sankhya 69:History 4831:  4792:  4784:  4686:  4624:JOSA A 4597:889822 4595:  4500:  4442:  4406:  4362:  3808:where 3710:does. 3477:where 851:, and 721:be a ( 701:, let 478:where 266:where 150:domain 59:metric 34:, the 4790:S2CID 4764:arXiv 4593:S2CID 4575:arXiv 4529:arXiv 4440:JSTOR 4404:S2CID 4829:ISSN 4782:ISSN 4684:ISSN 4360:ISBN 3876:and 3587:and 2709:Let 1413:and 765:are 745:and 507:and 425:and 368:For 128:and 105:For 77:, a 4860:doi 4821:doi 4774:doi 4676:doi 4585:doi 4539:doi 4525:536 4396:doi 4033:be 3441:det 3425:det 3414:det 1069:by 30:In 4879:: 4856:12 4854:. 4850:. 4827:. 4817:15 4815:. 4811:. 4788:. 4780:. 4772:. 4760:57 4758:. 4754:. 4714:, 4708:, 4682:. 4672:29 4670:. 4666:. 4591:. 4583:. 4571:55 4569:. 4537:. 4523:. 4519:. 4498:MR 4494:35 4492:. 4486:. 4474:^ 4464:. 4452:^ 4434:. 4430:. 4416:^ 4402:. 4392:46 4390:. 4386:. 4374:^ 4281:. 3722:: 3626:. 3545:. 3403:ln 3239:, 3073:ln 2773:, 365:. 217:ln 97:. 65:. 4866:. 4862:: 4835:. 4823:: 4796:. 4776:: 4766:: 4690:. 4678:: 4599:. 4587:: 4577:: 4545:. 4541:: 4531:: 4504:. 4446:. 4436:7 4410:. 4398:: 4368:. 4250:) 4247:Q 4244:, 4241:P 4238:( 4235:C 4232:B 4212:) 4209:Q 4206:, 4203:P 4200:( 4197:C 4194:B 4171:, 4164:i 4160:q 4154:i 4150:p 4142:n 4137:1 4134:= 4131:i 4123:= 4120:) 4116:q 4112:, 4108:p 4104:( 4101:C 4098:B 4073:i 4069:q 4046:i 4042:p 4021:i 4001:P 3981:n 3961:Q 3958:, 3955:P 3924:) 3921:X 3917:| 3913:1 3910:= 3907:Y 3904:( 3900:P 3896:= 3893:) 3890:X 3887:( 3862:) 3859:) 3856:X 3853:( 3844:1 3841:( 3838:) 3835:X 3832:( 3823:E 3819:= 3782:L 3771:2 3763:4 3757:1 3750:2 3747:1 3737:2 3734:1 3696:) 3693:q 3690:, 3687:p 3684:( 3681:C 3678:B 3672:1 3641:B 3637:D 3606:B 3602:D 3595:0 3575:1 3569:C 3566:B 3560:0 3529:. 3524:2 3518:2 3508:+ 3503:1 3490:= 3473:, 3460:) 3451:2 3435:1 3408:( 3398:2 3395:1 3390:+ 3387:) 3382:2 3367:1 3357:( 3352:1 3337:T 3333:) 3327:2 3312:1 3302:( 3297:8 3294:1 3289:= 3286:) 3281:2 3277:p 3273:, 3268:1 3264:p 3260:( 3255:B 3251:D 3227:) 3222:i 3211:, 3206:i 3196:( 3191:N 3186:= 3181:i 3177:p 3160:. 3147:) 3139:q 3129:p 3121:2 3114:2 3109:q 3101:+ 3096:2 3091:p 3080:( 3068:2 3065:1 3060:+ 3052:2 3047:q 3039:+ 3034:2 3029:p 3017:2 3013:) 3007:q 2994:p 2986:( 2978:4 2975:1 2970:= 2967:) 2964:q 2961:, 2958:p 2955:( 2950:B 2946:D 2920:2 2871:) 2866:2 2858:, 2852:( 2847:N 2825:) 2820:2 2815:q 2807:, 2802:q 2794:( 2789:N 2781:q 2761:) 2756:2 2751:p 2743:, 2738:p 2730:( 2725:N 2717:p 2689:] 2682:) 2679:x 2676:d 2673:( 2670:Q 2665:) 2662:x 2659:d 2656:( 2653:P 2646:[ 2640:Q 2636:E 2632:= 2629:) 2626:x 2623:d 2620:( 2617:Q 2610:) 2607:x 2604:d 2601:( 2598:Q 2593:) 2590:x 2587:d 2584:( 2581:P 2571:X 2562:= 2559:) 2556:x 2553:d 2550:( 2547:Q 2542:) 2539:x 2536:( 2533:p 2525:X 2516:= 2513:) 2510:Q 2507:, 2504:P 2501:( 2498:C 2495:B 2475:) 2472:x 2469:( 2463:) 2460:x 2457:d 2454:( 2451:Q 2446:) 2443:x 2440:d 2437:( 2434:P 2428:= 2425:) 2422:x 2419:( 2416:p 2396:Q 2376:) 2373:x 2370:d 2367:( 2364:Q 2361:) 2358:x 2355:( 2352:p 2349:= 2346:) 2343:x 2340:d 2337:( 2334:P 2314:1 2308:) 2305:Q 2302:, 2299:P 2296:( 2293:C 2290:B 2284:0 2244:) 2241:Q 2238:, 2235:P 2232:( 2229:C 2226:B 2214:. 2202:) 2199:x 2196:d 2193:( 2184:) 2181:x 2178:( 2175:q 2172:) 2169:x 2166:( 2163:p 2155:X 2146:= 2143:) 2140:Q 2137:, 2134:P 2130:| 2126:x 2123:d 2120:( 2117:c 2114:b 2108:X 2099:= 2096:) 2093:Q 2090:, 2087:P 2084:( 2081:C 2078:B 2063:. 2051:) 2048:x 2045:d 2042:( 2028:) 2025:x 2022:( 2015:q 2011:) 2008:x 2005:( 1998:p 1992:= 1989:) 1986:x 1983:d 1980:( 1971:) 1968:x 1965:( 1958:l 1954:) 1951:x 1948:( 1941:q 1937:) 1934:x 1931:( 1924:l 1920:) 1917:x 1914:( 1907:p 1901:= 1898:) 1895:x 1892:d 1889:( 1881:) 1878:x 1875:( 1872:l 1868:) 1865:x 1862:( 1859:q 1856:) 1853:x 1850:( 1847:l 1844:) 1841:x 1838:( 1835:p 1830:= 1827:) 1824:x 1821:( 1815:) 1812:x 1809:( 1806:l 1800:) 1797:x 1794:( 1791:q 1788:) 1785:x 1782:( 1779:p 1774:= 1771:) 1768:x 1765:d 1762:( 1753:) 1750:x 1747:( 1744:q 1741:) 1738:x 1735:( 1732:p 1727:= 1724:) 1721:Q 1718:, 1715:P 1711:| 1707:x 1704:d 1701:( 1698:c 1695:b 1672:Q 1660:, 1648:) 1645:x 1642:d 1639:( 1633:) 1630:x 1627:( 1620:l 1616:) 1613:x 1610:( 1603:p 1599:= 1596:) 1593:x 1590:d 1587:( 1581:) 1578:x 1575:( 1572:l 1569:) 1566:x 1563:( 1560:p 1557:= 1554:) 1551:x 1548:d 1545:( 1534:) 1531:x 1528:( 1521:p 1517:= 1514:) 1511:x 1508:d 1505:( 1499:) 1496:x 1493:( 1490:p 1487:= 1484:) 1481:x 1478:d 1475:( 1472:P 1446:) 1443:x 1440:( 1433:l 1429:= 1398:) 1395:x 1392:( 1389:l 1386:= 1275:. 1272:) 1269:x 1266:d 1263:( 1255:) 1252:x 1249:( 1243:) 1240:x 1237:d 1234:( 1226:) 1223:x 1220:d 1217:( 1214:Q 1208:) 1205:x 1202:( 1196:) 1193:x 1190:d 1187:( 1179:) 1176:x 1173:d 1170:( 1167:P 1159:= 1156:) 1153:x 1150:d 1147:( 1138:) 1135:x 1132:( 1129:q 1126:) 1123:x 1120:( 1117:p 1112:= 1109:) 1106:Q 1103:, 1100:P 1096:| 1092:x 1089:d 1086:( 1083:c 1080:b 1057:) 1052:B 1047:, 1042:X 1037:( 1017:) 1014:Q 1011:+ 1008:P 1005:( 999:2 996:1 990:= 927:q 924:, 921:p 901:) 898:x 895:d 892:( 886:) 883:x 880:( 877:q 874:= 871:) 868:x 865:d 862:( 859:Q 839:) 836:x 833:d 830:( 824:) 821:x 818:( 815:p 812:= 809:) 806:x 803:d 800:( 797:P 753:Q 733:P 689:) 684:B 679:, 674:X 669:( 649:Q 646:, 643:P 631:. 619:x 616:d 610:) 607:x 604:( 601:q 598:) 595:x 592:( 589:p 581:X 572:= 569:) 566:Q 563:, 560:P 557:( 554:C 551:B 524:) 521:x 518:( 515:q 495:) 492:x 489:( 486:p 466:x 463:d 460:) 457:x 454:( 451:q 448:= 445:) 442:x 439:d 436:( 433:Q 413:x 410:d 407:) 404:x 401:( 398:p 395:= 392:) 389:x 386:d 383:( 380:P 344:) 341:x 338:( 335:Q 332:) 329:x 326:( 323:P 314:X 306:x 298:= 295:) 292:Q 289:, 286:P 283:( 280:C 277:B 250:) 246:) 243:Q 240:, 237:P 234:( 231:C 228:B 224:( 211:= 208:) 205:Q 202:, 199:P 196:( 191:B 187:D 161:X 136:Q 116:P 20:)

Index

Bhattacharyya coefficient
statistics
similarity
probability distributions
statistical
metric
triangle inequality
Anil Kumar Bhattacharyya
statistician
Indian Statistical Institute
Sankhya
Indian Science Congress
Calcutta Mathematical Society
domain
probability density
sigma finite
absolutely continuous
normal distribution
multivariate normal
Mahalanobis distance
triangle inequality
Hellinger distance
Bayes error rate
classification
Mahalanobis distance
polar codes
speaker recognition
Bhattacharyya angle
Kullback–Leibler divergence
Hellinger distance

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.

↑