Better average - case predictions fork - binary
نویسنده
چکیده
Papers by Goldman, Rivest and Schapire, and Goldman and War-muth contain worst case analyses for several polynomial time algorithms for predicting k-binary relations. Here we present an empirical study of the average-case performance of those algorithms, and also of two new methods which dominate them in this average-case setting. The better of the two methods uses a graph coloring heuristic to solve for the target matrix row types. 1 The k-binary relation prediction problem In the k-brpp, an adversary chooses a hidden nm matrix of bits. Only k rows of this matrix are unique; the remaining n ? k rows are copies of these rows. Then, n m times, the adversary selects one of the covered bits in the matrix and asks us to predict its value. After guessing, we get to see the actual bit. One performance measure is the total number of mistakes that we make during the n m predictions for a given matrix. A more detailed measure is, for every t, the average accuracy of the prediction which occurs after t bits have been uncovered. In the average case experiments which are the subject of this paper, the adversary is a random process, i.e. the target matrices and the query orders are randomly generated. 1 2 Average case performance of earlier algorithms Suppose that part of the target matrix has been uncovered, and we have been asked to predict the bit which lies at matrix location (i; j). The only way to make a correct prediction is to predict the value at a location (i 0 ; j), where i 0 is a row which is of the same type as row i but which, unlike row i, is deened in column j (that is, where the entry in column j has already been uncovered). Unfortunately, we can rarely be sure that two rows of the partially uncovered matrix are of the same type. All of the diierent algorithms discussed in this paper amount to diierent methods for dealing with this uncertainty. Broadly 1 Speciically, we drew from a uniform distribution over all matrices with exactly n=k copies of each row type.
منابع مشابه
VLE Predictions of Strongly Non-Ideal Binary Mixtures by Modifying Van Der Waals and Orbey-Sandler Mixing Rules
By proposing a predictive method with no adjustable parameter and by using infinite dilution activity coefficients of components in binary mixtures obtained from UNIFAC model, the binary interaction parameters (k12) in van der Waals mixing rule (vdWMR) and Orbey-Sandler mixing rule (OSMR) have been evaluated. The predicted binary interaction parameters are used in Peng-Robinson-S...
متن کاملCorrelations and Predictions of THF + 2-Alkanol Binary Mixtures Behaviour by PC-SAFT Model and Friction Theory
In this article the behavior of tetrahydrofuran (THF) + 2-alkanol namely 2-propanol, 2-butanol, 2-pentanol, 2-hexanol and 2-heptanol binary mixtures through the density and viscosity measurements have been studied as a function of composition and within the temperature range of 293.15–313.15 K. The excess molar volume, isobaric thermal expansivity, partial molar volumes, and viscosity deviation...
متن کاملHow to say Greedy in Fork Algebras
Because of their expressive power, binary relations are widely used in program specification and development within formal calculi. The existence of a finite equational axiomatization for algebras of binary relations with a fork operation guarantees that the heuristic power coming from binary relations is captured inside an abstract equational calculus. In this paper we show how to express the ...
متن کاملBinary Wavelet Transform Based Histogram Feature for Content Based Image Retrieval
In this paper a new visual feature, binary wavelet transform based histogram (BWTH) is proposed for content based image retrieval. BWTH is facilitated with the color as well as texture properties. BWTH exhibits the advantages of binary wavelet transform and histogram. The performance of CBIR system with proposed feature is observed on Corel 1000 (DB1) and Corel 2450 (DB2) natural image database...
متن کاملA Comparative Study on the Forecast of Fresh Food Sales Using Logistic Regression, Moving Average and Bpnn Methods
Different prediction methods give different performance predictions when used for daily fresh food sales forecasting. Logistic Regression (LR) is a good choice for binary data, the Moving Average (MA) method is good for simple prediction, while the Back-Propagation Neural Network (BPNN) method is good for long term data. In this study we develop and compare the performance of three sales foreca...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2007