Machine learning has evolved into an enabling technology for a wide range of highly successful applications. The potential for this success to continue and accelerate has placed machine learning (ML) at the top of research, economic and political agendas. Such unprecedented interest is fuelled by a vision of ML applicability extending to healthcare, transportation, defence and other domains of great societal importance. Achieving this vision requires the use of ML in safety-critical applications that demand levels of assurance beyond those needed for current ML applications. Our paper provides a comprehensive survey of the state-of-the-art in the assurance of ML, i.e. in the generation of evidence that ML is sufficiently safe for its intended use. The survey covers the methods capable of providing such evidence at different stages of the machine learning lifecycle, i.e. of the complex, iterative process that starts with the collection of the data used to train an ML component for a system, and ends with the deployment of that component within the system. The paper begins with a systematic presentation of the ML lifecycle and its stages. We then define assurance desiderata for each stage, review existing methods that contribute to achieving these desiderata, and identify open challenges that require further research.

In this paper we survey the methods and concepts developed for the evaluation of dialogue systems. Evaluation is a crucial part during the development process. Often, dialogue systems are evaluated by means of human evaluations and questionnaires. However, this tends to be very cost and time intensive. Thus, much work has been put into finding methods, which allow to reduce the involvement of human labour. In this survey, we present the main concepts and methods. For this, we differentiate between the various classes of dialogue systems (task-oriented dialogue systems, conversational dialogue systems, and question-answering dialogue systems). We cover each class by introducing the main technologies developed for the dialogue systems and then by presenting the evaluation methods regarding this class.

No refund on tickets will be made under any circumstances unless otherwise instructed by the event organiser Venue Owner reserves the right without refund or compensation to refuse admission to any persons whose conduct is disorderly or unbecoming. The Promoter may add, withdraw or substitute artistes and/or vary advertised programmes, event times, seating arrangements and audience capacity without prior notice. The Promoter/Venue Owner may postpone, cancel,interrupt or stop the event due to adverse weather, dangerous situations, or any other causes beyond his reasonable control. Information is collected from those registering with www.showshappening.com in order to facilitate ticket purchase or other services available. This information is collected and used in accordance with ShowsHappening's Privacy Policy, which forms part of these conditions.

Do you dream of harnessing your engineering and machine learning knowledge to enrich users' lives? To improve their privacy and security? To expand their ecosystems by opening up models and data to the world? If so, you should join Mozilla's Machine Learning group as part of the three year, EU funded Bergamot Project. The goal of Bergamot is to extend Firefox with an open, on-device neural machine translation (NMT) engine, making translation local, private, and secure.

Multilingual representations have mostly been evaluated based on their performance on specific tasks. In this article, we look beyond engineering goals and analyze the relations between languages in computational representations. We introduce a methodology for comparing languages based on their organization of semantic concepts. We propose to conduct an adapted version of representational similarity analysis of a selected set of concepts in computational multilingual representations. Using this analysis method, we can reconstruct a phylogenetic tree that closely resembles those assumed by linguistic experts. These results indicate that multilingual distributional representations which are only trained on monolingual text and bilingual dictionaries preserve relations between languages without the need for any etymological information. In addition, we propose a measure to identify semantic drift between language families. We perform experiments on word-based and sentence-based multilingual models and provide both quantitative results and qualitative examples. Analyses of semantic drift in multilingual representations can serve two purposes: they can indicate unwanted characteristics of the computational models and they provide a quantitative means to study linguistic phenomena across languages. The code is available at https://github.com/beinborn/SemanticDrift.

Standard methods in deep learning for natural language processing fail to capture the compositional structure of human language that allows for systematic generalization outside of the training distribution. However, human learners readily generalize in this way, e.g. by applying known grammatical rules to novel words. Inspired by work in neuroscience suggesting separate brain systems for syntactic and semantic processing, we implement a modification to standard approaches in neural machine translation, imposing an analogous separation. The novel model, which we call Syntactic Attention, substantially outperforms standard methods in deep learning on the SCAN dataset, a compositional generalization task, without any hand-engineered features or additional supervision. Our work suggests that separating syntactic from semantic learning may be a useful heuristic for capturing compositional structure.

Models such as Sequence-to-Sequence and Image-to-Sequence are widely used in real world applications. While the ability of these neural architectures to produce variable-length outputs makes them extremely effective for problems like Machine Translation and Image Captioning, it also leaves them vulnerable to failures of the form where the model produces outputs of undesirable length. This behavior can have severe consequences such as usage of increased computation and induce faults in downstream modules that expect outputs of a certain length. Motivated by the need to have a better understanding of the failures of these models, this paper proposes and studies the novel output-size modulation problem and makes two key technical contributions. First, to evaluate model robustness, we develop an easy-to-compute differentiable proxy objective that can be used with gradient-based algorithms to find output-lengthening inputs. Second and more importantly, we develop a verification approach that can formally verify whether a network always produces outputs within a certain length. Experimental results on Machine Translation and Image Captioning show that our output-lengthening approach can produce outputs that are 50 times longer than the input, while our verification approach can, given a model and input domain, prove that the output length is below a certain size.

Scientists are constantly figuring out how to expand the field of use of this incredible invention, which enables computer software to progressively improve its actions by adopting knowledge gained from previous experience. Machine learning, also referred to as artificial intelligence due to its ability to perform tasks using its own judgment, has been the subject of both praise and controversy. However, the sophisticated algorithms that have served in providing you ads on social networks might have a grand future in philology, archaeology, and linguistics. According to Émilie Pagé-Perron, a Ph.D. candidate in Assyriology at the University of Toronto, we might be closer than we thought to deciphering numerous Middle-Eastern cuneiform tablets written in Sumerian and Akkadian languages, all of which are several thousand years old. Pagé-Perron is in charge of the project officially titled Machine Translation and Automated Analysis of Cuneiform Languages, which currently operates in Frankfurt, Toronto, and Los Angeles, using combined efforts to create a program capable of translating the clay tablets.

This work presents a joint solution to two challenging tasks: text generation from data and open information extraction. We propose to model both tasks as sequence-to-sequence translation problems and thus construct a joint neural model for both. Our experiments on knowledge graphs from Visual Genome, i.e., structured image analyses, shows promising results compared to strong baselines. Building on recent work on unsupervised machine translation, we report the first results - to the best of our knowledge - on fully unsupervised text generation from structured data.

Most machine translation systems generate text autoregressively, by sequentially predicting tokens from left to right. We, instead, use a masked language modeling objective to train a model to predict any subset of the target words, conditioned on both the input text and a partially masked target translation. This approach allows for efficient iterative decoding, where we first predict all of the target words non-autoregressively, and then repeatedly mask out and regenerate the subset of words that the model is least confident about. By applying this strategy for a constant number of iterations, our model improves state-of-the-art performance levels for constant-time translation models by over 3 BLEU on average. It is also able to reach 92-95% of the performance of a typical left-to-right transformer model, while decoding significantly faster.