Abstract

A radically new approach to statistical modelling, which combines mathematical techniques of Bayesian statistics with the philosophy of the theory of competitive on-line algorithms, has arisen over the last decade in computer science (to a large degree, under the influence of Dawid’s prequential statistics). In this approach, which we call “competitive on-line statistics”, it is not assumed that data are generated by some stochastic mechanism; the bounds derived for the performance of competitive on-line statistical procedures are guaranteed to hold (and not just hold with high probability or on the average). This paper reviews some results in this area; the new material in it includes the proofs for the performance of the Aggregating Algorithm in the problem of linear regression with square loss. Keywords: Bayes’s rule, competitive on-line algorithms, linear regression, prequential statistics, worst-case analysis.

Citations

2307	A decision-theoretic generalization of on-line learning and an application to boosting - Freund, Schapire - 1995
1686	An Introduction to Kolmogorov Complexity and its Applications - Li, Vitányi - 1993
678	The weighted majority algorithm - Littlestone, Warmuth - 1994
536	Applied Regression Analysis - Draper, Smith - 1981
404	A Formal Theory of Inductive Inference - Solomonoff - 1964
397	A universal prior for integers and estimation by minimum description length - Rissanen
317	How to use expert advice - Cesa-Bianchi, Freund, et al. - 1997
311	An Information Measure for Classification - Wallace, Boulton
251	Aggregating strategies - Vovk - 1990
247	Context-senstive learning methods for text categorization - CONNOR, Cohen, et al.
247	Exponentiated gradient versus gradient descent for linear predictors - Kivinen, Warmuth - 1997
198	Tracking the best expert - Herbster, Warmuth - 1995
189	Gambling in a rigged casino: The adversarial multi-arm bandit problem - Auer, Cesa-Bianchi, et al. - 1995
173	Statistical theory. The prequential approach (with discussion - Dawid - 1984
155	Universal portfolios - Cover - 1991
155	A dynamic disk spin-down technique for mobile computing, in - Helmbold, Long, et al. - 1996
134	Schapire, Adaptive game playing using multiplicative weights - Freund, E - 1999
133	Game theory, on-line prediction and boosting - Freund, Schapire - 1996
116	Relative loss bounds for online density estimation with the exponential family of distributions - Azoury, Warmuth
106	A game of prediction with expert advice - Vovk - 1998
93	Using and combining predictors that specialize - Freund, Schapire, et al. - 1997
85	Universal portfolios with side information - COVER, ORDENTLICH - 1996
75	Sequential prediction of individual sequences under general loss functions - Haussler, Kivinen, et al. - 1998
74	Tracking the best disjunction - Auer, Warmuth - 1998
72	The minimum description length principle and reasoning under uncertainty - Gru¨nwald - 1998
71	Predicting nearly as well as the best pruning of a decision tree - Helmbold, Schapire - 1997
58	Averaging expert predictions - Kivinen, Warmuth - 1999
55	Asymptotic minimax regret for data compression, gambling and prediction - Xie, Barron - 2000
55	Universal portfolios with and without transaction costs - Blum, Kalai - 1997
53	Tight worst-case loss bounds for predicting with expert advice - Haussler, Kivinen, et al. - 1995
49	A decision-theoretic extension of stochastic complexity and its applications to learning - Yamanishi - 1998
46	Derandomizing stochastic prediction strategies - Vovk - 1999
43	Probability forecasting - Dawid - 1986
40	Predicting a binary sequence almost as well as the optimal biased coin - Freund - 1996
36	On-line learning and the metrical task system problem - Blum, Burch
35	Learning probabilistic prediction functions - DeSantis, Markowski, et al. - 1988
31	Worstcase quadratic loss bounds for on-line prediction of linear functions by gradient descent - Cesa-Bianchi, Long, et al. - 1996
30	Stochastic complexity (with discussion - Rissanen - 1987
27	An efficient extension to mixture techniques for prediction and decision trees - Pereira, Singer - 1999
23	Fisherian inference in likelihood and prequential frames of reference (with discussion - Dawid - 1991
23	Machine-learning applications of algorithmic randomness - Vovk, Gammerman, et al. - 1999
22	C.: Universal portfolio selection - Vovk, Watkins - 1998
19	Prediction in the worst case - Foster - 1991
18	Tracking the best regressor - Herbster, Warmuth - 1998
17	A logic of probability, with application to the foundations of statistics (with discussion - Vovk
17	Competitive on-line linear regression - Vovk - 1998
15	Information theoretic asymptotics of Bayes methods - Clarke, Barron - 1990
14	On a randomness criterion - Vovk - 1987
13	Estimation and inference by compact coding (with discussions - WALLACE, FREEMAN - 1987
12	Universal forecasting algorithms - Vovk - 1992