Prediction markets provide an efficient means to assess uncertain quantities from forecasters. Traditional and competitive strictly proper scoring rules have been shown to incentivize players to provide truthful probabilistic forecasts. However, we show that when those players can cooperate, these mechanisms can instead discourage them from reporting what they really believe. When players with ...

Ensemble of networks has been proven to give better prediction result than a single network. Two commonly used way of determining the ensemble weights are simple average ensemble method and the generalized ensemble method. In the paper, we propose the weighted least square ensemble network. The major difference between this method and the other ensemble methods is that we do not assume that nei...

We point out a simple argument relying on hypercontractivity to describe tail inequalities on the distribution of the largest eigenvalues of random matrices at the rate given by the Tracy–Widom distribution. The result is illustrated on the known examples of the Gaussian and Laguerre unitary ensembles. The argument may be applied to describe the generic tail behavior of eigenfunction measures o...

We consider the combination of the outputs of several classifiers trained independently for the same discrimination task. We introduce new results which provide optimal solutions in the case of linear combinations. We compare our solutions to existing ensemble methods and characterize situations where our approach should be preferred.

Article history: Received 27 August 2012 Received in revised form 1 August 2013 Accepted 5 August 2013 Available online 15 August 2013

Abstract here...