The machine learning problem of classifier design is studied from the perspective of probability elicitation, in statistics. This shows that the standard approach of proceeding from the specification of a loss, to the minimization of conditional risk is overly restrictive. It is shown that a better alternative is to start from the specification of a functional form for the minimum conditional r...

One of the most challenging problems in computational advertising is the prediction of ad click and conversion rates for bidding in online advertising auctions. State-ofthe-art prediction methods include using the maximum entropy framework (also known as logistic regression) and log linear models. However, one unaddressed problem in the previous approaches is the existence of highly non-uniform...

We provide a PAC-Bayesian bound for the expected loss of convex combinations of classifiers under a wide class of loss functions (which includes the exponential loss and the logistic loss). Our numerical experiments with Adaboost indicate that the proposed upper bound, computed on the training set, behaves very similarly as the true loss estimated on the testing set.

Portfolio theory assumes that investors accept risk. This means thatin the equal rate of return on the two assets, the assets were chosenthat have a lower risk level. Modern portfolio theory is accepted byinvestors who believe that they are not cope with the market. Sothey keep many different types of securities in order to access theoptimum efficiency rate that is close to the rate of return o...