WrestleMania 33 is less than a month away, but there is still one more pay-per-view before WWE's biggest event of 2017. WWE Fastlane will feature the top wrestlers from "Monday Night Raw" at Milwaukee's Bradley Center Sunday night. The final PPV before WrestleMania is often uneventful, but that likely won't be the case this year. Not only are a few titles expected to change hands, but a couple of WWE superstars could make surprise returns. Goldberg will challenge Kevin Owens for the WWE Universal Championship, which Owens has held for more than six months.

This paper studies a new Bayesian algorithm for the joint reconstruction and classification of reflectance confocal microscopy (RCM) images, with application to the identification of human skin lentigo. The proposed Bayesian approach takes advantage of the distribution of the multiplicative speckle noise affecting the true reflectivity of these images and of appropriate priors for the unknown model parameters. A Markov chain Monte Carlo (MCMC) algorithm is proposed to jointly estimate the model parameters and the image of true reflectivity while classifying images according to the distribution of their reflectivity. Precisely, a Metropolis-whitin-Gibbs sampler is investigated to sample the posterior distribution of the Bayesian model associated with RCM images and to build estimators of its parameters, including labels indicating the class of each RCM image. The resulting algorithm is applied to synthetic data and to real images from a clinical study containing healthy and lentigo patients.

A secretive Canadian startup called Kindred AI is teaching robots how to perform difficult dexterous tasks at superhuman speeds by pairing them with human "pilots" wearing virtual-reality headsets and holding motion-tracking controllers. The technology offers a fascinating glimpse of how humans might work in synchronization with machines in the future, and it shows how tapping into human capabilities might amplify the capabilities of automated systems. For all the worry over robots and artificial intelligence eliminating jobs, there are plenty of things that machines still cannot do. The company demonstrated the hardware to MIT Technology Review last week, and says it plans to launch a product aimed at retailers in the coming months. The long-term ambitions are far grander.

The biggest holdup in making the boxing match between Conor McGregor and Floyd Mayweather a reality appears to be money. Both fighters are looking to make nine-figure paydays for what would undoubtedly become one of the highest-selling pay-per-views in history. Mayweather wants a guaranteed paycheck of $100 million to come out of retirement, and there is a precedent for the boxer earning such a large purse. It's how much he was paid to face Manny Pacquiao in 2015, and he eventually made more than twice that amount after receiving a cut of the PPV revenue. McGregor has asked for a $100 million payday, as well, but Mayweather isn't ready to agree to those terms. While McGregor is the UFC's biggest star, he's never had a purse of more than $3 million or even come close to approaching a nine-figure payday when factoring in his cut of the PPV revenue.

In this paper we introduce a new classification algorithm called Optimization of Distributions Differences (ODD). The algorithm aims to find a transformation from the feature space to a new space where the instances in the same class are as close as possible to one another while the gravity centers of these classes are as far as possible from one another. This aim is formulated as a multiobjective optimization problem that is solved by a hybrid of an evolutionary strategy and the Quasi-Newton method. The choice of the transformation function is flexible and could be any continuous space function. We experiment with a linear and a non-linear transformation in this paper. We show that the algorithm can outperform 6 other state-of-the-art classification methods, namely naive Bayes, support vector machines, linear discriminant analysis, multi-layer perceptrons, decision trees, and k-nearest neighbors, in 12 standard classification datasets. Our results show that the method is less sensitive to the imbalanced number of instances comparing to these methods. We also show that ODD maintains its performance better than other classification methods in these datasets, hence, offers a better generalization ability.

With WrestleMania 33 just a month away, it's been clear for a few weeks what WWE's top superstars will be doing at the biggest pay-per-view of 2017. A match between Brock Lesnar and Goldberg has been confirmed, while The Undertaker is likely to face Roman Reigns. The future of AJ Styles, however, hasn't been as clear. Aside from various rumors, the longest-reigning WWE Champion of the past year didn't have an obvious opponent for WrestleMania 33. The latest edition of "SmackDown Live," however, has provided the spark for Styles' next feud.

Modern retrieval systems are often driven by an underlying machine learning model. The goal of such systems is to identify and possibly rank the few most relevant items for a given query or context. Thus, such systems are typically evaluated using a ranking-based performance metric such as the area under the precision-recall curve, the $F_\beta$ score, precision at fixed recall, etc. Obviously, it is desirable to train such systems to optimize the metric of interest. In practice, due to the scalability limitations of existing approaches for optimizing such objectives, large-scale retrieval systems are instead trained to maximize classification accuracy, in the hope that performance as measured via the true objective will also be favorable. In this work we present a unified framework that, using straightforward building block bounds, allows for highly scalable optimization of a wide range of ranking-based objectives. We demonstrate the advantage of our approach on several real-life retrieval problems that are significantly larger than those considered in the literature, while achieving substantial improvement in performance over the accuracy-objective baseline.

Collaborative security initiatives are increasingly often advocated to improve timeliness and effectiveness of threat mitigation. Among these, collaborative predictive blacklisting (CPB) aims to forecast attack sources based on alerts contributed by multiple organizations that might be targeted in similar ways. Alas, CPB proposals thus far have only focused on improving hit counts, but overlooked the impact of collaboration on false positives and false negatives. Moreover, sharing threat intelligence often prompts important privacy, confidentiality, and liability issues. In this paper, we first provide a comprehensive measurement analysis of two state-of-the-art CPB systems: one that uses a trusted central party to collect alerts [Soldo et al., Infocom'10] and a peer-to-peer one relying on controlled data sharing [Freudiger et al., DIMVA'15], studying the impact of collaboration on both correct and incorrect predictions. Then, we present a novel privacy-friendly approach that significantly improves over previous work, achieving a better balance of true and false positive rates, while minimizing information disclosure. Finally, we present an extension that allows our system to scale to very large numbers of organizations.

The final pay-per-view before WrestleMania 33 is set for Sunday night in Milwaukee with WWE Fastlane 2017. The show will have major implications for WWE's biggest PPV of the year as multiple titles could change hands. Kevin Owens has held the WWE Universal Championship for six months, but he's in danger of losing the belt to Goldberg. Goldberg hasn't held a title in more than 13 years, but his return to WWE has gone so well that he appears to be headed for another championship run. While that match could help set up the main event for WrestleMania 33, the WrestleMania 33 Raw Women's Championship Match at the April 2 PPV could also be established at WWE Fastlane.

Discrimination-aware classification is receiving an increasing attention in the data mining and machine learning fields. The data preprocessing methods for constructing a discrimination-free classifier remove discrimination from the training data, and learn the classifier from the cleaned data. However, there lacks of a theoretical guarantee for the performance of these methods. In this paper, we fill this theoretical gap by mathematically bounding the probability that the discrimination in predictions is within a given interval in terms of the given training data and classifier. In our analysis, we adopt the causal model for modeling the mechanisms in data generation, and formally defining discrimination in the population, in a dataset, and in the prediction. The theoretical results show that the fundamental assumption made by the data preprocessing methods is not correct. Finally, we develop a framework for constructing a discrimination-free classifier with a theoretical guarantee.