On-line estimation with the multivariate Gaussian distribution
作者： Sanjoy Dasgupta , Daniel Hsu
DOI: 10.1007/978-3-540-72927-3_21
关键词: Multivariate kernel density estimation 、 Multivariate normal distribution 、 Gaussian function 、 Exponential family 、 Gaussian process 、 Applied mathematics 、 Covariance 、 Mathematics 、 Statistics 、 Gaussian random field 、 Density estimation
摘要: We consider on-line density estimation with the multivariate Gaussian distribution. In each of a sequence trials, learner must posit mean µ and covariance Σ; then receives an instance x incurs loss equal to negative log-likelihood under parameterized by (µ, Σ). prove bounds on regret for follow-the-leader strategy, which amounts choosing sample previously seen data.
springer.com
sci-hub.st 参考文章(13)
Eiji Takimoto, Manfred K. Warmuth, The Last-Step Minimax Algorithm Lecture Notes in Computer Science. pp. 279- 290 ,(2000) , 10.1007/3-540-40992-0_21
Katy S. Azoury, M. K. Warmuth, Relative loss bounds for on-line density estimation with the exponential family of distributions uncertainty in artificial intelligence. ,vol. 43, pp. 31- 40 ,(1999) , 10.1023/A:1010896012157
Elad Hazan, Adam Kalai, Satyen Kale, Amit Agarwal, Logarithmic Regret Algorithms for Online Convex Optimization Learning Theory. ,vol. 69, pp. 499- 513 ,(2006) , 10.1007/11776420_37
Koby Crammer, Online Tracking of Linear Subspaces Learning Theory. pp. 438- 452 ,(2006) , 10.1007/11776420_33
Manfred K. Warmuth, Eiji Takimoto, The Minimax Strategy for Gaussian Density Estimation. pp conference on learning theory. pp. 100- 106 ,(2000)
Nicolo Cesa-Bianchi, Gabor Lugosi, Prediction, learning, and games ,(2006)
Yoav Freund, Predicting a binary sequence almost as well as the optimal biased coin Proceedings of the ninth annual conference on Computational learning theory - COLT '96. pp. 89- 98 ,(1996) , 10.1145/238061.238072
Henri Frédéric Bohnenblust, Kenneth Joseph Arrow, Contributions to the theory of games ,(1950)
Shai Shalev-shwartz, Yoram Singer, Convex Repeated Games and Fenchel Duality neural information processing systems. pp. 1265- 1272 ,(2006)
Martin Zinkevich, Online convex programming and generalized infinitesimal gradient ascent international conference on machine learning. pp. 928- 935 ,(2003)