Adaboost is consistent

The risk, or probability of error, of the classifier produced by the AdaBoost algorithm is investigated. In particular, we consider the stopping strategy to be used in AdaBoost to achieve universal consistency. We show that provided AdaBoost is stopped after n1-ε iterations---for sample size n and ε ∈ (0,1)---the sequence of risks of the classifiers it produces approaches the Bayes risk.

A unifying view of multiple kernel learning

Recent research on multiple kernel learning has lead to a number of approaches for combining kernels in regularized risk minimization. The proposed approaches include different formulations of objectives and varying regularization strategies. In this paper we present a unifying general optimization criterion for multiple kernel learning and show how existing formulations are subsumed as special cases. We also derive the criterion's dual representation, which is suitable for general smooth optimization algorithms. Finally, we evaluate multiple kernel learning in this framework analytically using a Rademacher complexity bound on the generalization error and empirically in a set of experiments.

Matrix regularization techniques for online multitask learning

In this paper we examine the problem of prediction with expert advice in a setup where the learner is presented with a sequence of examples coming from different tasks. In order for the learner to be able to benefit from performing multiple tasks simultaneously, we make assumptions of task relatedness by constraining the comparator to use a lesser number of best experts than the number of tasks. We show how this corresponds naturally to learning under spectral or structural matrix constraints, and propose regularization techniques to enforce the constraints. The regularization techniques proposed here are interesting in their own right and multitask learning is just one application for the ideas. A theoretical analysis of one such regularizer is performed, and a regret bound that shows benefits of this setup is reported.

Closing the gap between bandit and full-information online optimization : high-probability regret bound

We demonstrate a modification of the algorithm of Dani et al for the online linear optimization problem in the bandit setting, which allows us to achieve an O( \sqrt{T ln T} ) regret bound in high probability against an adaptive adversary, as opposed to the in expectation result against an oblivious adversary of Dani et al. We obtain the same dependence on the dimension as that exhibited by Dani et al. The results of this paper rest firmly on those of Dani et al and the remarkable technique of Auer et al for obtaining high-probability bounds via optimistic estimates. This paper answers an open question: it eliminates the gap between the high-probability bounds obtained in the full-information vs bandit settings.

Adaptive online gradient descent

We study the rates of growth of the regret in online convex optimization. First, we show that a simple extension of the algorithm of Hazan et al eliminates the need for a priori knowledge of the lower bound on the second derivatives of the observed functions. We then provide an algorithm, Adaptive Online Gradient Descent, which interpolates between the results of Zinkevich for linear functions and of Hazan et al for strongly convex functions, achieving intermediate rates between [square root T] and [log T]. Furthermore, we show strong optimality of the algorithm. Finally, we provide an extension of our results to general norms.

Multitask learning with expert advice

We consider the problem of prediction with expert advice in the setting where a forecaster is presented with several online prediction tasks. Instead of competing against the best expert separately on each task, we assume the tasks are related, and thus we expect that a few experts will perform well on the entire set of tasks. That is, our forecaster would like, on each task, to compete against the best expert chosen from a small set of experts. While we describe the "ideal" algorithm and its performance bound, we show that the computation required for this algorithm is as hard as computation of a matrix permanent. We present an efficient algorithm based on mixing priors, and prove a bound that is nearly as good for the sequential task presentation case. We also consider a harder case where the task may change arbitrarily from round to round, and we develop an efficient approximate randomized algorithm based on Markov chain Monte Carlo techniques.

Learning the Kernel Matrix with Semi-Definite Programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive definite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space -- classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semi-definite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -- using the labelled part of the data one can learn an embedding also for the unlabelled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method to learn the 2-norm soft margin parameter in support vector machines, solving another important open problem. Finally, the novel approach presented in the paper is supported by positive empirical results.

The neurokinin 1 receptor antagonist, ezlopitant, reduces appetitive responding for sucrose and ethanol

Abstract Background: The current obesity epidemic is thought to be partly driven by over-consumption of sugar-sweetened diets and soft drinks. Loss-of-control over eating and addiction to drugs of abuse share overlapping brain mechanisms including changes in motivational drive, such that stimuli that are often no longer ‘liked’ are still intensely ‘wanted’ [7,8]. The neurokinin 1 (NK1) receptor system has been implicated in both learned appetitive behaviors and addiction to alcohol and opioids; however, its role in natural reward seeking remains unknown. Methodology/Principal Findings: We sought to determine whether the NK1-receptor system plays a role in the reinforcing properties of sucrose using a novel selective and clinically safe NK1-receptor antagonist, ezlopitant (CJ-11,974), in three animal models of sucrose consumption and seeking. Furthermore, we compared the effect of ezlopitant on ethanol consumption and seeking in rodents. The NK1-receptor antagonist, ezlopitant decreased appetitive responding for sucrose more potently than for ethanol using an operant self-administration protocol without affecting general locomotor activity. To further evaluate the selectivity of the NK1-receptor antagonist in decreasing consumption of sweetened solutions, we compared the effects of ezlopitant on water, saccharin-, and sodium chloride (NaCl) solution consumption. Ezlopitant decreased intake of saccharin but had no effect on water or salty solution consumption. Conclusions/Significance: The present study indicates that the NK1-receptor may be a part of a common pathway regulating the self-administration, motivational and reinforcing aspects of sweetened solutions, regardless of caloric value, and those of substances of abuse. Additionally, these results indicate that the NK1-receptor system may serve as a therapeutic target for obesity induced by over-consumption of natural reinforcers.

Neuronal nicotinic acetylcholine receptors as pharmacotherapeutic targets for the treatment of alcohol use disorders

Alcohol use disorders (AUDs) are complex and developing effective treatments will require the combination of novel medications and cognitive behavioral therapy approaches. Epidemiological studies have shown there is a high correlation between alcohol consumption and tobacco use, and the prevalence of smoking in alcoholics is as high as 80% compared to about 30% for the general population. Both preclinical and clinical data provide evidence that nicotine administration increases alcohol intake and nonspecific nicotinic receptor antagonists reduce alcohol-mediated behaviors. As nicotine interacts specifically with the neuronal nicotinic acetylcholine receptor (nAChR) system, this suggests that nAChRs play an important role in the behavioral effects of alcohol. In this review, we discuss the importance of nAChRs for the treatment of AUDs and argue that the use of FDA approved nAChR ligands, such as varenicline and mecamylamine, approved as smoking cessation aids may prove to be valuable treatments for AUDs. We also address the importance of combining effective medications with behavioral therapy for the treatment of alcohol dependent individuals.