In the past few years, Convolutional Neural Networks (CNNs) have been achieving state-of-the-art performance on a variety of problems. Many companies employ resources and money to generate these models and provide them as an API, therefore it is in their best interest to protect them, i.e., to avoid that someone else copies them. Recent studies revealed that state-of-the-art CNNs are vulnerable to adversarial examples attacks, and this weakness indicates that CNNs do not need to operate in the problem domain (PD). Therefore, we hypothesize that they also do not need to be trained with examples of the PD in order to operate in it. Given these facts, in this paper, we investigate if a target black-box CNN can be copied by persuading it to confess its knowledge through random non-labeled data. The copy is two-fold: i) the target network is queried with random data and its predictions are used to create a fake dataset with the knowledge of the network; and ii) a copycat network is trained with the fake dataset and should be able to achieve similar performance as the target network. This hypothesis was evaluated locally in three problems (facial expression, object, and crosswalk classification) and against a cloud-based API. In the copy attacks, images from both non-problem domain and PD were used. All copycat networks achieved at least 93.7% of the performance of the original models with non-problem domain data, and at least 98.6% using additional data from the PD. Additionally, the copycat CNN successfully copied at least 97.3% of the performance of the Microsoft Azure Emotion API. Our results show that it is possible to create a copycat CNN by simply querying a target network as black-box with random non-labeled data.

This paper develops a computationally efficient posterior sampling algorithm for Bayesian linear regression models with Gaussian errors. Our new approach is motivated by the fact that existing software implementations for Bayesian linear regression do not readily handle problems with large number of observations (hundreds of thousands) and predictors (thousands). Moreover, existing sampling algorithms for popular shrinkage priors are bespoke Gibbs samplers based on case-specific latent variable representations. By contrast, the new algorithm does not rely on case-specific auxiliary variable representations, which allows for rapid prototyping of novel shrinkage priors outside the conditionally Gaussian framework. Specifically, we propose a slice-within-Gibbs sampler based on the elliptical slice sampler of Murray et al. [2010].

Statisticians at German technical university Technische Universitat Dortmund built a model that used machine learning to predict Spain will win the 2018 World Cup. The prediction is based on 100,000 simulations of the tournament. Spain was followed by Germany, Brazil, France and Belgium in terms of their chances of winning. And it should be a good tournament because Spain, with a 17.8 percent chance of winning, is only slightly ahead of Germany at 17.1 percent. Brazil follows with 12.3 percent, and then it's France (11.2 percent) and Belgium (10.4 percent).

Parallel sentence extraction is a task addressing the data sparsity problem found in multilingual natural language processing applications. We propose a bidirectional recurrent neural network based approach to extract parallel sentences from collections of multilingual texts. Our experiments with noisy parallel corpora show that we can achieve promising results against a competitive baseline by removing the need of specific feature engineering or additional external resources. To justify the utility of our approach, we extract sentence pairs from Wikipedia articles to train machine translation systems and show significant improvements in translation performance.

Deep neural networks, and in particular recurrent networks, are promising candidates to control autonomous agents that interact in real-time with the physical world. However, this requires a seamless integration of temporal features into the network's architecture. For the training of and inference with recurrent neural networks, they are usually rolled out over time, and different rollouts exist. Conventionally, during inference the layers of a network are computed in a sequential manner resulting in sparse temporal integration of information and long response times. In this study, we present a theoretical framework to describe the set of all rollouts and demonstrate their differences in solving specific tasks. We prove that certain rollouts, also with only skip and no recurrent connections, enable earlier and more frequent responses, and show empirically that these early responses have better performance. The streaming rollout maximizes these properties and, in addition, enables a fully parallel execution of the network reducing the runtime on massively parallel devices. Additionally, we provide an open-source toolbox to design, train, evaluate, and online-interact with streaming rollouts.

One of the current issues in Brain-Computer Interface is how to deal with noisy Electroencephalography measurements organized as multidimensional datasets. On the other hand, recently, significant advances have been made in multidimensional signal completion algorithms that exploit tensor decomposition models to capture the intricate relationship among entries in a multidimensional signal. We propose to use tensor completion applied to EEG data for improving the classification performance in a motor imagery BCI system with corrupted measurements. Noisy measurements are considered as unknowns that are inferred from a tensor decomposition model. We evaluate the performance of four recently proposed tensor completion algorithms plus a simple interpolation strategy, first with random missing entries and then with missing samples constrained to have a specific structure (random missing channels), which is a more realistic assumption in BCI Applications. We measured the ability of these algorithms to reconstruct the tensor from observed data. Then, we tested the classification accuracy of imagined movement in a BCI experiment with missing samples. We show that for random missing entries, all tensor completion algorithms can recover missing samples increasing the classification performance compared to a simple interpolation approach. For the random missing channels case, we show that tensor completion algorithms help to reconstruct missing channels, significantly improving the accuracy in the classification of motor imagery, however, not at the same level as clean data. Tensor completion algorithms are useful in real BCI applications. The proposed strategy could allow using motor imagery BCI systems even when EEG data is highly affected by missing channels and/or samples, avoiding the need of new acquisitions in the calibration stage.

We describe a search-based approach to generating new levels for bullet hell games, which are action games characterized by and requiring avoidance of a very large amount of projectiles. Levels are represented using a domain-specific description language, and search in the space defined by this language is performed by a novel variant of the Map-Elites algorithm which incorporates a feasible- infeasible approach to constraint satisfaction. Simulation-based evaluation is used to gauge the fitness of levels, using an agent based on best-first search. The performance of the agent can be tuned according to the two dimensions of strategy and dexterity, making it possible to search for level configurations that require a specific combination of both. As far as we know, this paper describes the first generator for this game genre, and includes several algorithmic innovations.

The last ten years have been a rough time for many bank employees in Spain. The country's lenders have laid off 89,500 workers on the back of narrowing margins, industry consolidation, mass closures of branches and gathering digitization. In 2008, when the financial crisis struck, Spain was home to some 278,000 banking professionals; today there are just 195,000. Another 3,000 redundancies are expected in the coming months, as Santander and Bankia plan to further streamline their businesses, pushing the total number of layoffs close to 95,000. The job losses are unlikely to end there. In fact, they could accelerate, especially if a potential new threat to traditional branch and front-office jobs materializes: artificial intelligence (AI).

Is artificial intelligence the next hot thing in ETFs? One big-name investor seems to think so. On Friday, ETF Managers Group filed for the Rogers AI Global Macro ETF (BIKR), blending two popular elements in finance--Jim Rogers and artificial intelligence. BIKR will track an index of single-country ETFs that was developed by Ocean Capital Advisors, a company headed by Rogers, the famous commodity investor and author of several best-selling books on the topic. Rogers' Ocean Capital will also act as the sponsor of BIKR.

The random-forest technique has emerged in recent years as a powerful way to analyze large data sets while avoiding some of the pitfalls of other data-mining methods. It is based on the idea that some future event can be determined by a decision tree in which an outcome is calculated at each branch by reference to a set of training data. However, decision trees suffer from a well-known problem. In the latter stages of the branching process, decisions can become severely distorted by training data that is sparse and prone to huge variation at this kind of resolution, a problem known as overfitting. The random-forest approach is different.