We propose an algorithm called query by committee, in which a committee of students is trained on the same data set. The next query is chosen according to the principle of maximal disagreement. The algorithm is studied for two toy models: the high-low game and perceptron learning of another

as much as possible with the edge labels. That is, we want a clustering that maximizes the number of # edges within clusters, plus the number of edges between clusters (equivalently, minimizes the number of disagreements: the number of edges inside clusters plus the number of # edges between

This paper shows how a text classifier's need for labeled training data can be reduced by a combination of active learning and Expectation Maximization (EM) on a pool of unlabeled data. Query-by-Committee is used to actively select documents for labeling, then EM with a naive Bayes model

Two dierent problems are studied: (1) For an innite locally nite connected graph G, let p c (G) be the critical value for the existence of an ininte cluster in iid bond percolation on G and let P c = supfp c (G) : G transitive; p c (G) < 1g. Is P c < 1? (2) Let G be transitive with p c (G) < 1, let p 2 [0; 1] and let X and Y be iid bond percolations on G with retention parameters (1 + p)=2 and (1 p)=2 respectively. Is there a q < 1 such that p > q implies that for any monotone coupling ( ^ X ; ^ Y ) of X and Y the edges for which ^ X and ^ Y disagree form innite connected component(s) with positive probability? Let p d (G) be the inmum of such q's (including q = 1) and let P d = supfp d (G) : G transitive; p c (G) < 1g. Is the stronger statement P d < 1 true? On the other hand: Is it always true that p d (G) > p c (G)? It is shown that if one restricts attention to biregular planar graphs G then these two problems can be treated in a similar way and all the above questions are ...

We identify weakenings of the independence axiom of Nash (1950) to derive a solution without disagreement point. The idea is to determine disagreement points simultaneously with the solutions, and as a function of the utility space only. Our version of the Nash solution maximizes the Nash product w

Abstract- Studies on gender differences in realization patterns of disagreement in Chinese context, by so far, have been found insufficient. Disagreement is a commonly–occurred speech act in everyday human interactions and by its very nature is a face-threatening act that harms the solidarity

We consider the problem of finding a monomial (or a term) that maximizes the agreement rate with a given set of examples over the Boolean hypercube. The problem is motivated by learning of monomials in the agnostic framework of Haussler [12] and Kearns et al. [17]. Finding a monomial

In this paper a simulation to maximize the global solar radiation on a sloped collecting surface was applied to typical latitudes in the area of southern Italy, to calculate the optimum tilt angle of solar panels on building structures or large photovoltaic power plants located in that geographical

When designing a feeding program for gestating sows, we must remember the overall goals for the nutrition program: 1) prepare sows to be in proper body condition at farrowing; 2) maximize reproductive performance (farrowing rate and litter size); and 3) meet the daily nutrient requirements

seeks an estimated segmentation that is topologically equivalent and geometrically similar to the true, unknown segmentation. To this end, utilizing warping error as the evaluation metric, which measures topological disagreements between segmentations, the algorithm iteratively modifies the topology