Results 1  10
of
22
How to Use Expert Advice
 JOURNAL OF THE ASSOCIATION FOR COMPUTING MACHINERY
, 1997
"... We analyze algorithms that predict a binary value by combining the predictions of several prediction strategies, called experts. Our analysis is for worstcase situations, i.e., we make no assumptions about the way the sequence of bits to be predicted is generated. We measure the performance of the ..."
Abstract

Cited by 317 (66 self)
 Add to MetaCart
We analyze algorithms that predict a binary value by combining the predictions of several prediction strategies, called experts. Our analysis is for worstcase situations, i.e., we make no assumptions about the way the sequence of bits to be predicted is generated. We measure the performance of the algorithm by the difference between the expected number of mistakes it makes on the bit sequence and the expected number of mistakes made by the best expert on this sequence, where the expectation is taken with respect to the randomization in the predictions. We show that the minimum achievable difference is on the order of the square root of the number of mistakes of the best expert, and we give efficient algorithms that achieve this. Our upper and lower bounds have matching leading constants in most cases. We then show howthis leads to certain kinds of pattern recognition/learning algorithms with performance bounds that improve on the best results currently known in this context. We also compare our analysis to the case in which log loss is used instead of the expected number of mistakes.
Universal prediction of individual sequences
 IEEE Transactions on Information Theory
, 1992
"... AbstructThe problem of predicting the next outcome of an individual binary sequence using finite memory, is considered. The finitestate predictability of an infinite sequence is defined as the minimum fraction of prediction errors that can be made by any finitestate (FS) predictor. It is proved t ..."
Abstract

Cited by 158 (13 self)
 Add to MetaCart
AbstructThe problem of predicting the next outcome of an individual binary sequence using finite memory, is considered. The finitestate predictability of an infinite sequence is defined as the minimum fraction of prediction errors that can be made by any finitestate (FS) predictor. It is proved that this FS predictability can be attained by universal sequential prediction schemes. Specifically, an efficient prediction procedure based on the incremental parsing procedure of the LempelZiv data compression algorithm is shown to achieve asymptotically the FS predictability. Finally, some relations between compressibility and predictability are pointed out, and the predictability is proposed as an additional measure of the complexity of a sequence. Index TermsPredictability, compressibility, complexity, finitestate machines, Lempel Ziv algorithm.
Sequential Prediction of Individual Sequences Under General Loss Functions
 IEEE Transactions on Information Theory
, 1998
"... We consider adaptive sequential prediction of arbitrary binary sequences when the performance is evaluated using a general loss function. The goal is to predict on each individual sequence nearly as well as the best prediction strategy in a given comparison class of (possibly adaptive) prediction st ..."
Abstract

Cited by 75 (7 self)
 Add to MetaCart
We consider adaptive sequential prediction of arbitrary binary sequences when the performance is evaluated using a general loss function. The goal is to predict on each individual sequence nearly as well as the best prediction strategy in a given comparison class of (possibly adaptive) prediction strategies, called experts. By using a general loss function, we generalize previous work on universal prediction, forecasting, and data compression. However, here we restrict ourselves to the case when the comparison class is finite. For a given sequence, we define the regret as the total loss on the entire sequence suffered by the adaptive sequential predictor, minus the total loss suffered by the predictor in the comparison class that performs best on that particular sequence. We show that for a large class of loss functions, the minimax regret is either \Theta(log N) or \Omega\Gamma p ` log N ), depending on the loss function, where N is the number of predictors in the comparison class a...
Tracking the Best Disjunction
 Machine Learning
, 1995
"... . Littlestone developed a simple deterministic online learning algorithm for learning kliteral disjunctions. This algorithm (called Winnow) keeps one weight for each of the n variables and does multiplicative updates to its weights. We develop a randomized version of Winnow and prove bounds for a ..."
Abstract

Cited by 74 (11 self)
 Add to MetaCart
. Littlestone developed a simple deterministic online learning algorithm for learning kliteral disjunctions. This algorithm (called Winnow) keeps one weight for each of the n variables and does multiplicative updates to its weights. We develop a randomized version of Winnow and prove bounds for an adaptation of the algorithm for the case when the disjunction may change over time. In this case a possible target disjunction schedule T is a sequence of disjunctions (one per trial) and the shift size is the total number of literals that are added/removed from the disjunctions as one progresses through the sequence. We develop an algorithm that predicts nearly as well as the best disjunction schedule for an arbitrary sequence of examples. This algorithm that allows us to track the predictions of the best disjunction is hardly more complex than the original version. However the amortized analysis needed for obtaining worstcase mistake bounds requires new techniques. In some cases our low...
Tight WorstCase Loss Bounds for Predicting With Expert Advice
, 1994
"... this paper is somewhat different from the one just described. Assume that there are N experts E i , i = 1; : : : ; N , each trying to predict the outcomes y t as best they can. Let x t;i be the prediction of the ith expert E i about the ..."
Abstract

Cited by 53 (10 self)
 Add to MetaCart
this paper is somewhat different from the one just described. Assume that there are N experts E i , i = 1; : : : ; N , each trying to predict the outcomes y t as best they can. Let x t;i be the prediction of the ith expert E i about the
On Prediction of Individual Sequences
, 1998
"... Sequential randomized prediction of an arbitrary binary sequence is investigated. ..."
Abstract

Cited by 25 (5 self)
 Add to MetaCart
Sequential randomized prediction of an arbitrary binary sequence is investigated.
Combining forecasting procedures: some theoretical results
 Econometric Theory
, 2004
"... We study some methods of combining procedures for forecasting a continuous random variable. Statistical risk bounds under the square error loss are obtained under mild distributional assumptions on the future given the current outside information and the past observations. The risk bounds show that ..."
Abstract

Cited by 14 (2 self)
 Add to MetaCart
We study some methods of combining procedures for forecasting a continuous random variable. Statistical risk bounds under the square error loss are obtained under mild distributional assumptions on the future given the current outside information and the past observations. The risk bounds show that the combined forecast automatically achieves the best performance among the candidate procedures up to a constant factor and an additive penalty term. In term of the rate of convergence, the combined forecast performs as well as if one knew which candidate forecasting procedure is the best in advance. Empirical studies suggest combining procedures can sometimes improve forecasting accuracy compared to the original procedures. Risk bounds are derived to theoretically quantify the potential gain and price for linearly combining forecasts for improvement. The result supports the empirical finding that it is not automatically a good idea to combine forecasts. A blind combining can degrade performance dramatically due to the undesirable large variability in estimating the best combining weights. An automated combining method is shown in theory to achieve a balance between the potential gain and the complexity penalty (the price for combining); to take advantage (if any) of sparse combining; and to maintain the best performance (in rate) among the candidate forecasting procedures if linear or sparse combining does not help.
Agnostic Online Learning
"... We study learnability of hypotheses classes in agnostic online prediction models. The analogous question in the PAC learning model [Valiant, 1984] was addressed by Haussler [1992] and others, who showed that the VC dimension characterization of the sample complexity of learnability extends to the ag ..."
Abstract

Cited by 13 (2 self)
 Add to MetaCart
We study learnability of hypotheses classes in agnostic online prediction models. The analogous question in the PAC learning model [Valiant, 1984] was addressed by Haussler [1992] and others, who showed that the VC dimension characterization of the sample complexity of learnability extends to the agnostic (or ”unrealizable”) setting. In his influential work, Littlestone [1988] described a combinatorial characterization of hypothesis classes that are learnable in the online model. We extend Littlestone’s results in two aspects. First, while Littlestone only dealt with the realizable case, namely, assuming there exists a hypothesis in the class that perfectly explains the entire data, we derive results for the nonrealizable (agnostic) case as well. In particular, we describe several models of nonrealizable data and derive upper and lower bounds on the achievable regret. Second, we extend the theory to include marginbased hypothesis classes, in which the prediction of each hypothesis is accompanied by a confidence value. We demonstrate how the newly developed theory seamlessly yields novel online regret bounds for the important class of large margin linear separators. 1
Prediction & Information Theory
"... n the same way, based on what the subject has seen in the past. Then the text is completely recoverable from the sequence consisting of the numbers of guesses: 1; 1; 1; 5; 1; 1; 2;:::;1; 1; 1; 1; 1 #1# For example, the subject looks at the #rst integer in this sequence and sees a #1". The subject t ..."
Abstract

Cited by 9 (0 self)
 Add to MetaCart
n the same way, based on what the subject has seen in the past. Then the text is completely recoverable from the sequence consisting of the numbers of guesses: 1; 1; 1; 5; 1; 1; 2;:::;1; 1; 1; 1; 1 #1# For example, the subject looks at the #rst integer in this sequence and sees a #1". The subject then knows that the #rst letter of the text is the letter that the subject would make as his#her #rst guess as the #rst letter of any text. Presumably, the subject has a rule that tells him#her to always make the initial guess for the #rst letter of any text to be #T". The #rst letter of the text is therefore decoded correctly. One can arrive at a data compression algorithm based on Shannon's idea. If x 1 ;x 2 ;:::;:::;x i,1 represents the #rst i , 1 letters