The TriRhenaTech alliance universities and their partners presented their competences in the field of artificial intelligence and their cross-border cooperations with the industry at the tri-national conference 'Artificial Intelligence : from Research to Application' on March 13th, 2019 in Offenburg. The TriRhenaTech alliance is a network of universities in the Upper Rhine Trinational Metropolitan Region comprising of the German universities of applied sciences in Furtwangen, Kaiserslautern, Karlsruhe, and Offenburg, the Baden-Wuerttemberg Cooperative State University Loerrach, the French university network Alsace Tech (comprised of 14 'grandes \'ecoles' in the fields of engineering, architecture and management) and the University of Applied Sciences and Arts Northwestern Switzerland. The alliance's common goal is to reinforce the transfer of knowledge, research, and technology, as well as the cross-border mobility of students.

Researchers in neural machine translation (NMT) and natural language processing (NLP) may want to keep an eye on a new framework from Google. Lingvo is specifically tailored toward sequence models and NLP, which includes speech recognition, language understanding, MT, and speech translation. The Google AI team claims there are already "dozens" of research papers in these areas based on Lingvo. In fact, they said this was one reason they decided to open-source the project: to support the research community and encourage reproducible results. Lingvo supports multiple neural network architectures -- from recurrent neural nets to Transformer models -- and comes with lots of documentation on common implementations across different tasks (i.e., NLP, NMT, speech synthesis).

Multilingual Neural Machine Translation (NMT) models are capable of translating between multiple source and target languages. Despite various approaches to train such models, they have difficulty with zero-shot translation: translating between language pairs that were not together seen during training. In this paper we first diagnose why state-of-the-art multilingual NMT models that rely purely on parameter sharing, fail to generalize to unseen language pairs. We then propose auxiliary losses on the NMT encoder that impose representational invariance across languages. Our simple approach vastly improves zero-shot translation quality without regressing on supervised directions. For the first time, on WMT14 English-FrenchGerman, we achieve zero-shot performance that is on par with pivoting. We also demonstrate the easy scalability of our approach to multiple languages on the IWSLT 2017 shared task.

In this article I describe a research agenda for securing machine learning models against adversarial inputs at test time. This article does not present results but instead shares some of my thoughts about where I think that the field needs to go. Modern machine learning works very well on I.I.D. data: data for which each example is drawn {\em independently} and for which the distribution generating each example is {\em identical}. When these assumptions are relaxed, modern machine learning can perform very poorly. When machine learning is used in contexts where security is a concern, it is desirable to design models that perform well even when the input is designed by a malicious adversary. So far most research in this direction has focused on an adversary who violates the {\em identical} assumption, and imposes some kind of restricted worst-case distribution shift. I argue that machine learning security researchers should also address the problem of relaxing the {\em independence} assumption and that current strategies designed for robustness to distribution shift will not do so. I recommend {\em dynamic models} that change each time they are run as a potential solution path to this problem, and show an example of a simple attack using correlated data that can be mitigated by a simple dynamic defense. This is not intended as a real-world security measure, but as a recommendation to explore this research direction and develop more realistic defenses.

The well-known Gumbel-Max trick for sampling from a categorical distribution can be extended to sample $k$ elements without replacement. We show how to implicitly apply this 'Gumbel-Top-$k$' trick on a factorized distribution over sequences, allowing to draw exact samples without replacement using a Stochastic Beam Search. Even for exponentially large domains, the number of model evaluations grows only linear in $k$ and the maximum sampled sequence length. The algorithm creates a theoretical connection between sampling and (deterministic) beam search and can be used as a principled intermediate alternative. In a translation task, the proposed method compares favourably against alternatives to obtain diverse yet good quality translations. We show that sequences sampled without replacement can be used to construct low-variance estimators for expected sentence-level BLEU score and model entropy.

This paper describes a baseline for the second iteration of the Fact Extraction and VERification shared task (FEVER2.0) which explores the resilience of systems through adversarial evaluation. We present a collection of simple adversarial attacks against systems that participated in the first FEVER shared task. FEVER modeled the assessment of truthfulness of written claims as a joint information retrieval and natural language inference task using evidence from Wikipedia. A large number of participants made use of deep neural networks in their submissions to the shared task. The extent as to whether such models understand language has been the subject of a number of recent investigations and discussion in literature. In this paper, we present a simple method of generating entailment-preserving and entailment-altering perturbations of instances by common patterns within the training data. We find that a number of systems are greatly affected with absolute losses in classification accuracy of up to $29\%$ on the newly perturbed instances. Using these newly generated instances, we construct a sample submission for the FEVER2.0 shared task. Addressing these types of attacks will aid in building more robust fact-checking models, as well as suggest directions to expand the datasets.

The world's biggest consumer electronics show was held last month and wandering around the seemingly endless stalls of emerging new products, it was impossible to avoid the claims of artificial intelligence in some form or another. Some gadgets were, of course, smarter than others. From facial recognition food bowls for your pets to handheld speech recognition and language translation devices, smart tech and self-learning algorithms abound. The actual intelligence of some smart products is debatable but the trend is undeniable.Source:Supplied Encompassing terms including deep learning, machine learning, neural networks and general artificial intelligence which seeks to build computers with a capacity to think and learn like humans, it can be hard to pin down what AI truly means. But it's clearly here to stay.

It has been shown that PE can not only yield productivity gains of 36% [9], but that it also increases the quality [2]. This paper discusses how human and artificial intelligence can be combined for efficient language translations, which tools already exist and which open challenges remain (see Figure 1). HARNESSING SYNERGIES BETWEEN AIS AND HUMANS Draft Proposal The PE process starts with an initial draft that is proposed by the AI and which the human uses as a basis. There are two main sources for this proposal: a machine translation (MT) and a translation memory (TM). Simply put, TMs are large databases containing already completed human translations which are matched (using fuzzy or exact matches) against the sentence to be translated to provide a starting point for PE. Machines can easily generate a variety of probable translations from (a combination of) MT and TM instead of only a single one; however, proposing too many and maybe even highly similar translations could overwhelm the human.

The repeated reuse of test sets in popular benchmark problems raises doubts about the credibility of reported test error rates. Verifying whether a learned model is overfitted to a test set is challenging as independent test sets drawn from the same data distribution are usually unavailable, while other test sets may introduce a distribution shift. We propose a new hypothesis test that uses only the original test data to detect overfitting. It utilizes a new unbiased error estimate that is based on adversarial examples generated from the test data and importance weighting. Overfitting is detected if this error estimate is sufficiently different from the original test error rate. The power of the method is illustrated using Monte Carlo simulations on a synthetic problem. We develop a specialized variant of our dependence detector for multiclass image classification, and apply it to testing overfitting of recent models to two popular real-world image classification benchmarks. In the case of ImageNet, our method was not able to detect overfitting to the test set for a state-of-the-art classifier, while on CIFAR-10 we found strong evidence of overfitting for the two recent model architectures we considered, and weak evidence of overfitting on the level of individual training runs.

A year ago we told you why Quality Estimation is the missing piece in Machine Translation. Today, we have some exciting news to share about a new project from our AI Research team, with my colleagues Fábio Kepler, Sony Trénous, and Miguel Vera. Since 2016, Unbabel's AI team has been focused on advancing the state of the art in Quality Estimation (QE). Our models are running in production systems for 14 language pairs, with coverage and performance improving over time, thanks to the increasing amount of data produced by our human post-editors on a daily basis. This combination of AI and humans is what makes our translation pipeline fast and accurate, at scale.