Predicting values in sequence
作者： Rina Panigrahy , Mikhail Kapralov
DOI:
关键词: Multiple data 、 Data prediction 、 Confidence value 、 Data value 、 Statistics 、 Sequence (medicine) 、 Computer science 、 Value (computer science)
摘要: Multiple data prediction strategies are received. Each strategy may predict a next value in sequence of values with corresponding confidence value. Rather than rely on single strategy, the predictions each linearly combined to generate that is more accurate and has lower overall loss any individual strategies. Further, deviation calculated based have been observed so far using weighted sum favors recent over less sequence. A generated deviation.
lens.org 参考文章(26)
Stephen D. Bay, Combining Nearest Neighbor Classifiers Through Multiple Feature Subsets international conference on machine learning. pp. 37- 45 ,(1998)
Stefan Zemke, Data mining for prediction - financial series case Data- och systemvetenskap. ,(2003)
David Haussler, Jyrki Kivinen, Manfred K. Warmuth, Tight worst-case loss bounds for predicting with expert advice european conference on computational learning theory. pp. 69- 83 ,(1994) , 10.1007/3-540-59119-2_169
Kai Ming Ting, Boon Toh Low, Model Combination in the Multiple-Data-Batches Scenario european conference on machine learning. pp. 250- 265 ,(1997) , 10.1007/3-540-62858-4_90
Elad Hazan, Amit Agarwal, Efficient Algorithms for Online Game Playing and Universal Portfolio Management. Electronic Colloquium on Computational Complexity. ,vol. 13, ,(2006)
Eyal Even-Dar, Michael Kearns, Yishay Mansour, Jennifer Wortman, Regret to the best vs. regret to the average conference on learning theory. pp. 233- 247 ,(2007) , 10.1007/978-3-540-72927-3_18
Elad Hazan, Satyen Kale, Extracting certainty from uncertainty: regret bounded by variation in costs Machine Learning. ,vol. 80, pp. 165- 188 ,(2010) , 10.1007/S10994-010-5175-X
Thomas P. Hayes, Varsha Dani, Robbing the bandit: less regret in online geometric optimization against an adaptive adversary symposium on discrete algorithms. pp. 937- 943 ,(2006) , 10.5555/1109557.1109660
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
H. Brendan McMahan, Adam Tauman Kalai, Abraham D. Flaxman, Online convex optimization in the bandit setting: gradient descent without a gradient symposium on discrete algorithms. pp. 385- 394 ,(2005) , 10.5555/1070432.1070486